Full text of "Year book - Carnegie Institution of Washington"

See other formats

Digitized by the Internet Archive

in 2012 with funding from

LYRASIS Members and Sloan Foundation

http://www.archive.org/details/yearbookcarne52195253carn

CARNEGIE INSTITUTION
OF WASHINGTON

YEAR BOOK No. 52
July 1, 1952— June 30, 1953

With Administrative Reports through December 11, 1953

erf ^

CARNEGIE INSTITUTION OF WASHINGTON
WASHINGTON, D. C.

1953

THE LORD BALTIMORE PRESS, BALTIMORE, MARYLAND

CONTENTS

PAGE

Officers and Staff 1953 v

Organization, Plan, and Scope xi

Articles of Incorporation xii

By-Laws of the Institution xv

Abstract of Minutes of the Fifty-Fifth Meeting of the Board of Trustees .... xxi

Report of the Executive Committee xxiii

Report of Auditors xxv

Report of the President 1-18

Reports of Departmental Activities and Co-operative Studies

Astronomy

Mount Wilson and Palomar Observatories 3

Terrestrial Sciences

Geophysical Laboratory 39

Department of Terrestrial Magnetism 97

Biological Sciences

Department of Plant Biology 145

Department of Embryology 183

Department of Genetics 205

Archaeology

Department of Archaeology 249

Reports of Research Associates

E. A. Lowe 297

Bibliography 299

Index 301

111

PRESIDENT AND TRUSTEES

James F. Bell
Robert Woods Bliss
Lindsay Bradford
Omar N. Bradley
W. Cameron Forbes
Walter S. Gifford
Crawford H. Greenewalt
Caryl P. Haskins

Robert Woods Bliss
Vannevar Bush

Alfred L. Loomis

Barklie Henry

Robert A. Lovett

Crawford H. Greenewalt

Barklie Henry

Alfred L. Loomis

James F. Bell

PRESIDENT
Vannevar Bush

BOARD OF TRUSTEES

Elihu Root, Jr., Chairman

Lindsay Bradford, Vice-Chairman

Robert Woods Bliss, Secretary *

Barklie Henry
Ernest O. Lawrence
Alfred L. Loomis
Robert A. Lovett
Keith S. McHugh
Roswell Miller
Henry S. Morgan
Seeley G. Mudd

Executive Committee

Elihu Root, Jr., Chairman
Caryl P. Haskins
Barklie Henry
Henry S. Morgan

Finance Committee

Lindsay Bradford, Chairman
Henry S. Morgan
Henning W. Prentis, Jr.

Auditing Committee
Roswell Miller, Chairman

Nominating Committee

Henry S. Morgan, Chairman
Keith S. McHugh

Committee on Astronomy

Seeley G. Mudd, Chairman
Roswell Miller

Committee on Terrestrial Sciences
Ernest O. Lawrence, Chairman

Committee on Biological Sciences

Caryl P. Haskins, Chairman
William I. Myers

Committee on Archaeology

Henry R. Shepley, Chairman
Robert Woods Bliss

William I. Myers
Henning W. Prentis, Jr.
David Rockefeller
Elihu Root, Jr.
Henry R. Shepley
Charles P. Taft
Juan T. Trippe
Robert E. Wilson

Henning W. Prentis, Jr.
Henry R. Shepley

Elihu Root, Jr.

Keith S. McHugh

Elihu Root, Jr.

David Rockefeller

Charles P. Taft

Juan T. Trippe

* Secretary of Board of Trustees March 26 — December 11, 1953, Henry S. Morgan.

FORMER PRESIDENTS AND TRUSTEES

PRESIDENTS

Daniel Coit Gilman, 1902-1904 Robert Simpson Woodward, 1904-1920

John Campbell Merriam, President 1 921-1938; President Emeritus 1 939-1945

TRUSTEES

Alexander Agassiz
George J. Baldwin
Thomas Barbour
John S. Billings
Robert S. Brookings
John L. Cadwalader
William W. Campbell
John J. Carty
Whitefoord R. Cole
Frederic A. Delano
Cleveland H. Dodge
William E. Dodge
Charles P. Fenner
Homer L. Ferguson
Simon Flexner
James Forrestal
William N. Frew
Lyman J. Gage
Cass Gilbert
Frederick H. Gillett
Daniel C. Gilman
John Hay
Myron T. Herrick
Abram S. Hewitt
Henry L. Higginson
Ethan A. Hitchcock
Henry Hitchcock
Herbert Hoover
William Wirt FIowe
Charles L. Hutchinson
Walter A. Jessup
Frank B. Jewett
Samuel P. Langley
Charles A. Lindbergh
William Lindsay
Henry Cabot Lodge

1904-05
1925-27
1934-46
1902-13
1910-29
1903-14
1929-38
1916-32

J 9 2 5-34
1927-49

1903-23

1902-03

1914-24

1927-52

1910-14

1948-49

1902-15

1902-12

1924-34

I 9 2 4"35
1902-08

1902-05

1915-29

1902-03

1902-19

1902-09

1902-02

1920-49

1903-09

1902-04

1938-44

!933-49
1904-06

T 934"39
1902-09

1914-24

Seth Low
Wayne MacVeagh
Andrew W. Mellon
Darius O. Mills
S. Weir Mitchell
Andrew J. Montague
William W. Morrow
William Church Osborn
James Parmelee
Wm. Barclay Parsons
Stewart Paton
George W. Pepper
John J. Pershing
Henry S. Pritchett
Gordon S. Rentschler
Elihu Root
Julius Rosenwald
Martin A. Ryerson
Theobald Smith
John C. Spooner
William Benson Storey
Richard P. Strong
William H. Taft
William S. Thayer
James W. Wadsworth
Charles D. Walcott
Frederic C. Walcott
Henry P. Walcott
Lewis H. Weed
William H. Welch
Andrew D. White
Edward D. White
Henry White
George W. Wickersham
Robert S. Woodward
Carroll D. Wright

1902-16
1902-07
1924-37
1902-09
1902-14

1902-29
1927-34
1917-31
1907-32
1916-42
1914-19

I 93°~43
1906-36

1946-48

1902-37

1929-31

1908-28

1914-34

1902-07

1924-39

1934-48

1906-15

1929-32

1932-52

1902-27

1931-48

1910-24

J935-5 2
1906-34

1902-03

1913-27

1909-36

1905-24

1902-08

Besides the names enumerated above, the following were ex-officio members of the Board
of Trustees under the original charter, from the date of organization until April 28, 1904:
the President of the United States, the President of the Senate, the Speaker of the House
of Representatives, the Secretary of the Smithsonian Institution, the President of the
National Academy of Sciences.

STAFF OF INVESTIGATORS FOR THE YEAR 1953

ASTRONOMY

Mount Wilson and Palomar Observatories

813 Santa Barbara Street, Pasadena 4, California

Mount Wilson Observatory organized in 1904; George E. Hale, Director 1 904-1 923, Honorary Director
1923-1936; Walter S. Adams, Director 1924-1945. Unified operation with the Palomar Observatory of
the California Institute of Technology began in 1948.

Ira S. Bowen, Director
Walter Baade
Horace W. Babcock
William A. Baum
Armin J. Deutsch
Jesse L. Greenstein
|Edwin P. Hubble
Milton L. Humason
Rudolph L. Minkowski

Guido Munch
Seth B. Nicholson
Donald E. Osterbrock
Edison Pettit
Robert S. Richardson
Allan R. Sandage
# Albert G. Wilson
Olin C. Wilson
Fritz Zwicky

TERRESTRIAL SCIENCES

Geophysical Laboratory

2801 Upton Street, N.W., Washington 8, D. C.

Organized in 1906, opened in 1907; Arthur L. Day, Director 1909-1936; Leason H. Adams, Acting
Director 1 936-1 937, Director 1938 — August 1, 1952; George W. Morey, Acting Director August 1, 1952 —
August 31, 1953.

Philip H. Abelson, Director

Francis R. Boyd, Jr.

Felix Chayes

Gordon L. Davis

Joseph L. England

Hans P. Eugster

Joseph W. Greig

Department of Terrestrial Magnetism

Frank C. Kracek
William S. MacKenzie
George W. Morey
J. Frank Schairer
# 0. Frank Tuttle
Hatten S. Yoder, Jr.

5241 Broad Branch Road, N.W., Washington 15, D. C.

Organized in 1904; Louis A. Bauer, Director 1 904-1 929; John A. Fleming, Acting Director 1 929-1 934,
Director 1 935-1 946.

Merle A. Tuve, Director
JPhilip H. Abelson

L. Thomas Aldrich

Ellis T. Bolton

Roy J. Britten

Bernard F. Burke

Dean B. Cowie

John W. Firor

Scott E. Forbush

John W. Graham

Norman P. Heydenburg
§Ellis A. Johnson

Richard B. Roberts

Howard E. Tatel

Georges M. Temmer

George R. Tilton

Ernest H. Vestine

Harry W. Wells

George W. Wetherill

Visiting Investigators
William R. Duryee
||George B. Field
Paul W. Gast
Pembroke J. Hart
H. Lawrence Helfer
B. Y. Mills
Leif Owren
Irena Z. Roberts
F. Graham Smith
M. Sugiura

* Resigned in 1953.

t Deceased in 1953.

tTo August 31, 1953.

§ On leave of absence.

|| Term of appointment completed in 1953.

Vll

CARNEGIE INSTITUTION OF WASHINGTON

BIOLOGICAL SCIENCES
Department of Plant Biology

Stanford, California

Desert Laboratory, opened in 1903, became headquarters of Department of Botanical Research in 1905;
name changed to Laboratory for Plant Physiology in 1923; Daniel T. MacDougal, Director 1906-1927.
Reorganized in 1928 as Division of Plant Biology, including Ecology; Herman A. Spoehr, Chairman
1 927-1 930 and 1 931-1947, Chairman Emeritus 1 947-1 950. Name changed to Department of Plant
Biology in 1951.

C. Stacy French, Director Guest Investigators
Jens C. Clausen Robert Hill

Paul Grun Hiroshi Tamiya

William M. Hiesey Hemming I. Virgin

Harold W. Milner

Malcolm A. Nobs Visiting Investigators
James H. C. Smith Elias Landolt

Violet K. Young Jack E. Myers

Department of Embryology

Wolfe and Madison Streets, Baltimore 5, Maryland
Organized in 191 4; Franklin P. Mall, Director 191 4-19 17; George L. Streeter, Director 191 8-1940

*George W. Corner, Director Consultant

George W. Bartelmez

Samuel R. M. Reynolds, Acting Director
David W. Bishop
Bent G. Boving
Robert K. Burns
Arpad Csapo

Research Associate
Elizabeth M. Ramsey

Department of Genetics

Cold Spring Harbor, hong Island, New Yor\

Station for Experimental Evolution opened in 1904; name changed to Department of Experimental
Evolution in 1906; combined with Eugenics Record Office in 1921 to form Department of Genetics.
Charles B. Davenport, Director 1904-1934; Albert F. Blakeslee, Director 1935-1941.

Milislav Demerec, Director
Alfred D. Hershey
Berwind P. Kaufmann
Barbara McClintock
Margaret R. McDonald
JN. Visconti

§EVELYN M. WlTKIN

* On leave of absence to September 1, 1953.

t Resigned in 1953.

§ On leave of absence.

|| Term of appointment completed in 1953.

Special Investigators
Maurice H. Bernstein

||A. M. M. Berrie

1 1 Irving Galinsky
Alan Garen
Jean Hemmerly
June Dixon Hudis

||Edgar L. Labrum
Dan L. Lindsley
Hermann Moser
Neville Symonds

||r. c. von borstel

Vlll

STAFF OF INVESTIGATORS FOR THE YEAR 1953

ARCHAEOLOGY

Department of Archaeology

io Frisbie Place, Cambridge 38, Massachusetts

Department of Historical Research organized in 1903; Andrew C. McLaughlin, Director 1 903-1 905;
J. Franklin Jameson, Director 1 905-1 928. In 1930 this Department was incorporated as a section of
United States history in a new Division of Historical Research; Alfred V. Kidder, Chairman 1 930-1 950.
Name changed to Department of Archaeology in 1951.

Harry E. D. Pollock, Director Edwin M. Shook

Margaret W. Harrison, Editor A. Ledyard Smith

Earl H. Morris Robert E. Smith

Tatiana Proskouriakoff Gustav Stromsvik

Karl Ruppert J. Eric S. Thompson
Anna O. Shepard

RESEARCH ASSOCIATES

Research Associates Engaged in Post-Retirement Studies
Chester H. Heuser, Embryology Ralph L. Roys, History

Research Associates Connected with Other Institutions

Ralph W. Chaney (University of California), Paleobotany
Louis B. Flexner (University of Pennsylvania), Embryology
Arthur T. Hertig (Boston Lying-in Hospital), Embryology
E. A. Lowe (The Institute for Advanced Study), Paleography

OFFICE OF ADMINISTRATION

Vannevar Bush, President

Paul A. Scherer, Executive Officer

Samuel Callaway, Assistant to the President

Ailene J. Bauer, Director of Publications

Dorothy R. Swift, Editor

John A. Fleming, Adviser in International Scientific Relations

Earle B. Biesecker, Bursar

J. Stanley Lingebach, Assistant Bursar

James F. Sullivan, Assistant to the Bursar

Richard F. F. Nichols, Executive Assistant to the Finance Committee

ORGANIZATION, PLAN, AND SCOPE

The Carnegie Institution of Washington was founded by Andrew Carnegie,
January 28, 1902, when he gave to a board of trustees an endowment of registered
bonds of the par value of ten million dollars. To this fund an addition of two
million dollars was made by Mr. Carnegie on December 10, 1907, and a further
addition of ten million dollars was made by him on January 19, 191 1. Further-
more, the income of a reserve fund of about three million dollars, accumulated
in accordance with the founder's specifications in 191 1, is now available for general
use, and in recent years a total of ten million dollars has been paid by the Carnegie
Corporation of New York as increase to the Endowment Fund of the Institution.
The Institution was originally organized under the laws of the District of Columbia
and incorporated as the Carnegie Institution, articles of incorporation having been
executed on January 4, 1902. The Institution was reincorporated, however, by an
act of the Congress of the United States, approved April 28, 1904, under the title
o£ Carnegie Institution of Washington. (See existing Articles of Incorporation on
following pages.)

Organization under the new Articles of Incorporation was effected May 18, 1904,
and the Institution was placed under the control of a board of twenty-four trustees,
all of whom had been members of the original corporation. The trustees meet
annually in December to consider the affairs of the Institution in general, the progress
of work already undertaken, and the initiation of new projects, and to make
the necessary appropriations for the ensuing year. During the intervals between
the meetings of the trustees the affairs of the Institution are conducted by an
Executive Committee chosen by and from the Board of Trustees and acting through
the President of the Institution as chief executive officer.

The Articles of Incorporation of the Institution declare in general "that the
objects of the corporation shall be to encourage, in the broadest and most liberal
manner, investigation, research, and discovery, and the application of knowledge
to the improvement of mankind."

The Institution is essentially an operating organization. It attempts to advance
fundamental research in fields not normally covered by the activities of other agencies,
and to concentrate its attention upon specific problems, with the idea of shifting
attack from time to time to meet the more pressing needs of research as they develop
with increase of knowledge. Some of these problems require the collaboration of
several investigators, special equipment, and continuous effort. Many close relations
exist among activities of the Institution, and a type of organization representing
investigations in astronomy, in terrestrial sciences, in biological sciences, and in
archaeology has been effected. Conference groups on various subjects have played
a part in bringing new vision and new methods to bear upon many problems.
Constant efforts are made to facilitate interpretation and application of results of
research activities of the Institution.

ARTICLES OF INCORPORATION

Public No. 260. An Act to incorporate the Carnegie Institution of Washington.

Be it enacted by the Senate and House of Representatives of the United States of
America in Congress assembled, That the persons following, being persons who are
now trustees of the Carnegie Institution, namely, Alexander Agassiz, John S. Billings,
John L. Cadwalader, Cleveland H. Dodge, William N. Frew, Lyman J. Gage,
Daniel C. Gilman, John Hay, Henry L. Higginson, William Wirt Howe, Charles L.
Hutchinson, Samuel P. Langley, William Lindsay, Seth Low, Wayne MacVeagh,
Darius O. Mills, S. Weir Mitchell, William W. Morrow, Ethan A. Hitchcock,
Elihu Root, John C. Spooner, Andrew D. White, Charles D. Walcott, Carroll D.
Wright, their associates and successors, duly chosen, are hereby incorporated and
declared to be a body corporate by the name of the Carnegie Institution of Wash-
ington and by that name shall be known and have perpetual succession, with the
powers, limitations, and restrictions herein contained.

Sec. 2. That the objects of the corporation shall be to encourage, in the broadest
and most liberal manner, investigation, research, and discovery, and the application
of knowledge to the improvement of mankind; and in particular —

(a) To conduct, endow, and assist investigation in any department of science,
literature, or art, and to this end to cooperate with governments, universities, colleges,
technical schools, learned societies, and individuals.

(b) To appoint committees of experts to direct special lines of research.

(d) To conduct lectures, hold meetings, and acquire and maintain a library.

(e) To purchase such property, real or personal, and construct such building or
buildings as may be necessary to carry on the work of the corporation.

(f) In general, to do and perform all things necessary to promote the objects
of the institution, with full power, however, to the trustees hereinafter appointed
and their successors from time to time to modify the conditions and regulations
under which the work shall be carried on, so as to secure the application of the
funds in the manner best adapted to the conditions of the time, provided that the
objects of the corporation shall at all times be among the foregoing or kindred thereto.

Sec. 3. That the direction and management of the affairs of the corporation and
the control and disposal of its property and funds shall be vested in a board of trustees,
twenty-two in number, to be composed of the following individuals: Alexander
Agassiz, John S. Billings, John L. Cadwalader, Cleveland H. Dodge, William N.
Frew, Lyman J. Gage, Daniel C. Gilman, John Hay, Henry L. Higginson, William
Wirt Howe, Charles L. Hutchinson, Samuel P. Langley, William Lindsay, Seth
Low, Wayne MacVeagh, Darius O. Mills, S. Weir Mitchell, William W. Morrow,
Ethan A. Hitchcoc\, Elihu Root, John C. Spooner, Andrew D. White, Charles D.
Walcott, Carroll D. Wright, who shall constitute the first board of trustees. The
board of trustees shall have power from time to time to increase its membership
to not more than twenty-seven members. Vacancies occasioned by death, resignation,
or otherwise shall be filled by the remaining trustees in such manner as the by-laws

xii

ARTICLES OF INCORPORATION

shall prescribe; and the persons so elected shall thereupon become trustees and
also members of the said corporation. The principal place of business of the said
corporation shall be the city of Washington, in the District of Columbia.

Sec. 4. That such board of trustees shall be entitled to take, hold, and administer
the securities, funds, and property so transferred by said Andrew Carnegie to the
trustees of the Carnegie Institution and such other funds or property as may at any
time be given, devised, or bequeathed to them, or to such corporation, for the purposes
of the trust; and with full power from time to time to adopt a common seal, to
appoint such officers, members of the board of trustees or otherwise, and such
employees as may be deemed necessary in carrying on the business of the corporation,
at such salaries or with such remuneration as they may deem proper; and with
full power to adopt by-laws from time to time and such rules or regulations as
may be necessary to secure the safe and convenient transaction of the business of
the corporation; and with full power and discretion to deal with and expend the
income of the corporation in such manner as in their judgment will best promote
the objects herein set forth and in general to have and use all powers and authority
necessary to promote such objects and carry out the purposes of the donor. The
said trustees shall have further power from time to time to hold as investments
the securities hereinabove referred to so transferred by Andrew Carnegie, and any
property which has been or may be transferred to them or such corporation by
Andrew Carnegie or by any other person, persons, or corporation, and to invest
any sums or amounts from time to time in such securities and such form and
manner as are permitted to trustees or to charitable or literary corporations for
investment, according to the laws of the States of New York, Pennsylvania, or
Massachusetts, or in such securities as are authorized for investment by the said deed
of trust so executed by Andrew Carnegie, or by any deed of gift or last will and
testament to be hereafter made or executed.

Sec. 5. That the said corporation may take and hold any additional donations,
grants, devises, or bequests which may be made in further support of the purposes
of the said corporation, and may include in the expenses thereof the personal expenses
which the trustees may incur in attending meetings or otherwise in carrying out
the business of the trust, but the services of the trustees as such shall be gratuitous.

Sec. 6. That as soon as may be possible after the passage of this Act a meeting
of the trustees hereinbefore named shall be called by Daniel C. Gilman, John S.
Billings, Charles D. Walcott, S. Weir Mitchell, John Hay, Elihu Root, and Carroll D.
Wright, or any four of them, at the city of Washington, in the District of Columbia,
by notice served in person or by mail addressed to each trustee at his place of resi-
dence; and the said trustees, or a majority thereof, being assembled, shall organize
and proceed to adopt by-laws, to elect officers and appoint committees, and generally
to organize the said corporation; and said trustees herein named, on behalf of the
corporation hereby incorporated, shall thereupon receive, take over, and enter into
possession, custody, and management of all property, real or personal, of the cor-
poration heretofore known as the Carnegie Institution, incorporated, as hereinbefore
set forth under "An Act to establish a Code of Law for the District of Columbia,
January fourth, nineteen hundred and two," and to all its rights, contracts, claims,
and property of any kind or nature; and the several officers of such corporation, or

xiii

CARNEGIE INSTITUTION OF WASHINGTON

any other person having charge of any of the securities, funds, real or personal,
books, or property thereof, shall, on demand, deliver the same to the said trustees
appointed by this Act or to the persons appointed by them to receive the same;
and the trustees of the existing corporation and the trustees herein named shall
and may take such other steps as shall be necessary to carry out the purposes of
this Act.

Sec. 7. That the rights of the creditors of the said existing corporation known as
the Carnegie Institution shall not in any manner be impaired by the passage of this
Act, or the transfer of the property hereinbefore mentioned, nor shall any liability
or obligation for the payment of any sums due or to become due, or any claim or
demand, in any manner or for any cause existing against the said existing corporation,
be released or impaired; but such corporation hereby incorporated is declared to
succeed to the obligations and liabilities and to be held liable to pay and discharge
all of the debts, liabilities, and contracts of the said corporation so existing to the
same effect as if such new corporation had itself incurred the obligation or liability
to pay such debt or damages, and no such action or proceeding before any court
or tribunal shall be deemed to have abated or been discontinued by reason of the
passage of this Act.

Sec. 8. That Congress may from time to time alter, repeal, or modify this Act
of incorporation, but no contract or individual right made or acquired shall thereby
be divested or impaired.

Sec 9. That this Act shall take effect immediately.

Approved, April 28, 1904

XIV

BY-LAWS OF THE INSTITUTION

Adopted December 13, 1904. Amended December 13, 1910, December 13, 1912, December 10, 1937,
December 15, 1939, December 13, 1940, December 18, 1942, and December 12, 1947

Article I

THE TRUSTEES

1. The Board of Trustees shall consist of twenty-four members, with power to
increase its membership to not more than twenty-seven members. The Trustees shall
hold office continuously and not for a stated term.

2. In case any Trustee shall fail to attend three successive annual meetings of the
Board he shall thereupon cease to be a Trustee.

3. No Trustee shall receive any compensation for his services as such.

4. All vacancies in the Board of Trustees shall be filled by the Trustees by ballot
at an annual meeting, but no person shall be declared elected unless he receives the
votes of two-thirds of the Trustees present.

Article II

OFFICERS OF THE BOARD

i. The officers of the Board shall be a Chairman of the Board, a Vice-Chairman,
and a Secretary, who shall be elected by the Trustees, from the members of the
Board, by ballot to serve for a term of three years. All vacancies shall be filled by the
Board for the unexpired term; provided, however, that the Executive Committee
shall have power to fill a vacancy in the office of Secretary to serve until the next
meeting of the Board of Trustees.

2. The Chairman shall preside at all meetings and shall have the usual powers of a
presiding officer.

3. The Vice-Chairman, in the absence or disability of the Chairman, shall perform
the duties of the Chairman.

4. The Secretary shall issue notices of meetings of the Board, record its transactions,
and conduct that part of the correspondence relating to the Board and to his duties.

Article III

EXECUTIVE ADMINISTRATION

The President

1. There shall be a President who shall be elected by ballot by, and hold office during
the pleasure of, the Board, who shall be the chief executive officer of the Institution.
The President, subject to the control of the Board and the Executive Committee,
shall have general charge of all matters of administration and supervision of all
arrangements for research and other work undertaken by the Institution or with its
funds. He shall prepare and submit to the Board of Trustees and to the Executive

CARNEGIE INSTITUTION OF WASHINGTON

Committee plans and suggestions for the work of the Institution, shall conduct its
general correspondence and the correspondence with applicants for grants and with
the special advisers of the Committee, and shall present his recommendations in each
case to the Executive Committee for decision. All proposals and requests for grants
shall be referred to the President for consideration and report. He shall have power
to remove, appoint, and, within the scope of funds made available by the Trustees,
provide for compensation of subordinate employees and to fix the compensation of
such employees within the limits of a maximum rate of compensation to be
established from time to time by the Executive Committee. He shall be ex officio a
member of the Executive Committee.

2. He shall be the legal custodian of the seal and of all property of the Institution
whose custody is not otherwise provided for. He shall sign and execute on behalf
of the corporation all contracts and instruments necessary in authorized administrative
and research matters and affix the corporate seal thereto when necessary, and may
delegate the performance of such acts and other administrative duties in his absence
to the Executive Officer. He may execute all other contracts, deeds, and instruments
on behalf of the corporation and affix the seal thereto when expressly authorized by
the Board of Trustees or Executive Committee. He may, within the limits of his
own authorization, delegate to the Executive Officer authority to act as custodian of
and affix the corporate seal. He shall be responsible for the expenditure and disburse-
ment of all funds of the Institution in accordance with the directions of the Board
and of the Executive Committee, and shall keep accurate accounts of all receipts and
disbursements. Following approval by the Executive Committee he shall transmit
to the Board of Trustees before its annual meeting in December a written report of
the operations and business of the Institution for the preceding fiscal year with his
recommendations for work and appropriations for the succeeding calendar year.

3. He shall attend all meetings of the Board of Trustees.

4. There shall be an officer designated Executive Officer who shall be appointed by
and hold office at the pleasure of the President, subject to the approval of the
Executive Committee. His duties shall be to assist and act for the President as the
latter may duly authorize and direct.

5. The President shall retire from office at the end of the calendar year in which
he becomes sixty-five years of age.

Article IV

MEETINGS

1. The annual meeting of the Board of Trustees shall be held in the City of Wash-
ington, in the District of Columbia, on the first Friday following the second Thursday
of December in each year unless the date and place of meeting are otherwise ordered
by the Executive Committee.

2. Special meetings of the Board may be called by the Executive Committee by
notice served personally upon, or mailed to the usual address of, each Trustee twent)
days prior to the meeting.

3. Special meetings shall, moreover, be called in the same manner by the Chairman
upon the written request of seven members of the Board.

xvi

BY-LAWS OF THE INSTITUTION
Article V

COMMITTEES

i. There shall be the following standing Committees, viz. an Executive Committee,
a Finance Committee, an Auditing Committee, and a Nominating Committee.

2. All vacancies occurring in the Executive Committee, the Finance Committee,
the Auditing Committee, and the Nominating Committee shall be filled by the
Trustees at the next regular meeting. In case of vacancy in the Finance Committee,
the Auditing Committee, or the Nominating Committee, upon request of the re-
maining members of such committee, the Executive Committee may fill such vacancy
by appointment until the next meeting of the Board of Trustees.

3. The terms of all officers and of all members of committees, as provided for
herein, shall continue until their successors are elected or appointed.

Executive Committee

4. The Executive Committee shall consist of the Chairman and Secretary of the
Board of Trustees and the President of the Institution ex officio and, in addition,
five trustees to be elected by the Board by ballot for a term of three years, who shall
be eligible for re-election. Any member elected to fill a vacancy shall serve for the
remainder of his predecessor's term.

5. The Executive Committee shall, when the Board is not in session and has not
given specific directions, have general control of the administration of the affairs of
the corporation and general supervision of all arrangements for administration,
research, and other matters undertaken or promoted by the Institution. It shall also
submit to the Board of Trustees a printed or typewritten report of each of its meetings,
and at the annual meeting shall submit to the Board a report for publication.

6. The Executive Committee shall have power to authorize the purchase, sale,
exchange, or transfer of real estate.

Finance Committee

7. The Finance Committee shall consist of five members to be elected by the Board
of Trustees by ballot for a term of three years.

8. The Finance Committee shall have custody of the securities of the corporation
and general charge of its investments and invested funds, and shall care for and
dispose of the same subject to the directions of the Board of Trustees. It shall have
power to authorize the purchase, sale, exchange, or transfer of securities and to
delegate this power. It shall consider and recommend to the Board from time to
time such measures as in its opinion will promote the financial interests of the
Institution, and shall make a report at each meeting of the Board.

Auditing Committee

9. The Auditing Committee shall consist of three members to be elected by the
Board of Trustees by ballot for a term of three years.

xvii

CARNEGIE INSTITUTION OF WASHINGTON

10. Before each annual meeting of the Board of Trustees, the Auditing Com-
mittee shall cause the accounts of the Institution for the preceding fiscal year to be
audited by public accountants. The accountants shall report to the Committee, and
the Committee shall present said report at the ensuing annual meeting of the Board
with such recommendations as the Committee may deem appropriate.

No m in ating Coram ittee

ii. The Nominating Committee shall consist of the Chairman of the Board of
Trustees ex officio and, in addition, three trustees to be elected by the Board by
ballot for a term of three years, who shall not be eligible for re-election until after
the lapse of one year. Any member elected to fill a vacancy shall serve for the
remainder of his predecessor's term, provided that of the Nominating Committee
first elected after adoption of this By-Law one member shall serve for one year, one
member shall serve for two years, and one member shall serve for three years, the
Committee to determine the respective terms by lot.

12. Sixty days prior to an annual meeting of the Board the Nominating Com-
mittee shall notify the Trustees by mail of the vacancies to be filled in membership
of the Board. Each Trustee may submit nominations for such vacancies. Nomina-
tions so submitted shall be considered by the Nominating Committee, and ten days
prior to the annual meeting the Nominating Committee shall submit to members of
the Board by mail a list of the persons so nominated, with its recommendations for
filling existing vacancies on the Board and its Standing Committees. No other
nominations shall be received by the Board at the annual meeting except with the
unanimous consent of the Trustees present.

Article VI

FINANCIAL ADMINISTRATION

i. No expenditure shall be authorized or made except in pursuance of a previous
appropriation by the Board of Trustees, or as provided in Article V, paragraph 8,
hereof.

2. The fiscal year of the Institution shall commence on the first day of July in
each year.

3. The Executive Committee shall submit to the annual meeting of the Board a
full statement of the finances and work of the Institution for the preceding fiscal
year and a detailed estimate of the expenditures of the succeeding calendar year.

4. The Board of Trustees, at the annual meeting in each year, shall make general
appropriations for the ensuing calendar year; but nothing contained herein shall
prevent the Board of Trustees from making special appropriations at any meeting.

5. The Executive Committee shall have general charge and control of all appropria-
tions made by the Board. Following the annual meeting each year, the Executive
Committee may make allotment of funds for the period from January 1 to termi-
nation of the fiscal year on June 30. It may also make allotment of funds for the
period from July 1 to December 31 in advance of July 1. The Committee shall,
however, have full authority for allotment of available funds to meet necessary

xviii

BY-LAWS OF THE INSTITUTION

expenditures by other methods, if desirable, and transfer of balances to meet
special needs. It shall make provision for outstanding obligations and for revertment
of unexpended balances at termination of the fiscal year.

6. The securities of the Institution and evidences of property, and funds invested
and to be invested, shall be deposited in such safe depository or in the custody of
such trust company and under such safeguards as the Finance Committee shall
designate, subject to directions of the Board of Trustees. Income of the Institution
available for expenditure shall be deposited in such banks or depositories as may
from time to time be designated by the Executive Committee.

7. Any trust company entrusted with the custody of securities by the Finance
Committee may, by resolution of the Board of Trustees, be made Fiscal Agent of the
Institution, upon an agreed compensation, for the transaction of the business coming
within the authority of the Finance Committee.

Article VII

AMENDMENT OF BY-LAWS

i. These by-laws may be amended at any annual or special meeting of the Board
of Trustees by a two-thirds vote of the members present, provided written notice of
the proposed amendment shall have been served personally upon, or mailed to the
usual address of, each member of the Board twenty days prior to the meeting.

XIX

ABSTRACT OF MINUTES OF THE FIFTY-FIFTH MEETING OF

THE BOARD OF TRUSTEES

The annual meeting of the Board of Trustees was held in Washington, D. C, on
Friday, December n, 1953. The Chairman, Mr. Root, presided.

The following Trustees were in attendance: James F. Bell, Robert Woods Bliss,
Lindsay Bradford, Omar N. Bradley, W. Cameron Forbes, Crawford H. Greenewalt,
Caryl P. Haskins, Barklie Henry, Ernest O. Lawrence, Alfred L. Loomis, Robert A.
Lovett, Keith S. McHugh, Roswell Miller, Henry S. Morgan, William I. Myers,
Henning W. Prentis, Jr., David Rockefeller, Elihu Root, Jr., Henry R. Shepley,
Charles P. Taft, and Juan T. Trippe. The President of the Institution, Vannevar
Bush, was present.

The minutes of the fifty-fourth meeting were approved.

Reports of the President, the Executive Committee, the Finance Committee, the
Auditor, the Auditing Committee, and the Nominating Committee, and of the
Directors of Departments and Research Associates of the Institution were presented
and considered.

The sum of $1,847,584 was appropriated for the calendar year 1954 for expendi-
ture under the general charge and control of the Executive Committee.

The Chairman referred to the death of Lewis H. Weed almost a year ago and
also to the earlier action of the Board in holding open a position in its membership
pending the completion by Mr. Gifford of his duties as Ambassador to the Court
of St. James's. Walter S. Gifford, of New York, N. Y., was re-elected a member of
the Board to fill the vacancy held open for him, and Robert E. Wilson, of Chicago,
Illinois, was elected a member of the Board to fill the other vacancy.

Robert Woods Bliss was elected Secretary of the Board for the remainder of the
term ending in 1954 to succeed Henry S. Morgan, who had been serving under an
interim appointment. Caryl P. Haskins and Barklie Henry were re-elected members
of the Executive Committee for three-year terms, and Henry S. Morgan was elected
a member of the Executive Committee for the remainder of the term of Mr. Bliss,
expiring in 1954, as the Secretaryship carries with it ex-officio membership in the
Executive Committee. Lindsay Bradford was re-elected Chairman and member of
the Finance Committee to serve for three years, and Henning W. Prentis, Jr., was
re-elected a member of the Finance Committee to serve for three years. Keith S.
McHugh was elected a member of the Nominating Committee for a period of three
years, succeeding Robert Woods Bliss.

XXI

REPORT OF THE EXECUTIVE COMMITTEE
For the Year Ended June 30, 1953

To the Trustees of the Carnegie Institution of Washington:

Gentlemen: In accordance with the provisions of the By-Laws, the Executive
Committee submits this report to the annual meeting of the Board of Trustees.

During the fiscal year ended June 30, 1953, the Executive Committee held four
meetings, printed reports of which have been mailed to each Trustee and constitute
a part of this report.

The detailed record of the activities of the Institution is presented in the reports
from the Departments, which are contained in the Year Book, a review of some of
the highlights being given in the report of the President. The estimate of expendi-
tures for the calendar year 1954 contained in the report of the President has been
considered and approved by the Executive Committee, and the Committee has also
provisionally approved and recommends to the Board the proposed budget based
thereon.

The Board of Trustees, at its meeting of December 12, 1952, appointed the firm of
Haskins & Sells to audit the accounts of the Institution for the fiscal year ending June
30, 1953. The report of the Auditor, including a balance sheet showing assets and
liabilities of the Institution on June 30, 1953, together with supporting statements and
schedules, is submitted as a part of the report of the Executive Committee.

Two vacancies exist in the membership of the Board of Trustees: one vacancy left
unfilled since the resignation of Mr. Walter S. Gifford in 1950, and the other resulting
from the death of Dr. Lewis H. Weed on December 21, 1952.

A vacancy in the office of Secretary of the Board of Trustees, resulting from the
death of Dr. Weed, has been filled by the Executive Committee through the tem-
porary appointment of Mr. Morgan, until the forthcoming annual meeting of the
Board of Trustees. The terms of Dr. Haskins and Mr. Henry as members of the
Executive Committee will end at the annual meeting. The terms of Mr. Bradford
(Chairman) and of Mr. Prentis as members of the Finance Committee will also end
at the annual meeting. The term of Mr. Bliss as a member of the Nominating Com-
mittee will likewise end at the annual meeting.

Elihu Root, Jr., Chairman
October 22, ig^

xxin

HASKINS & SELLS 500 Equitable Building

Certified Public Accountants Baltimore 2

ACCOUNTANTS' CERTIFICATE

To the Board of Trustees of Carnegie Institution of Washington:

We have examined the balance sheet of Carnegie Institution of Washington as of
June 30, 1953 and the related summaries of current income and expenditures, current
funds surplus, current restricted gifts and grants, changes in endowment and other
special funds, and changes in investment in real estate and equipment for the year
then ended (Exhibits A to F, inclusive). Our examination was made in accordance
with generally accepted auditing standards, and included such tests of the accounting
records and such other auditing procedures as we considered necessary in the circum-
stances, except that we did not examine the records in support of expenditures made
(approximately $266,000) by five of the seven branch offices of the Institution, but we
have reviewed internal audit reports of the Bursar's office covering examinations of
all branch records during the year.

In our opinion, subject to the exception stated above with respect to the limitation
of the scope of our examination, the accompanying balance sheet and above described
summaries (Exhibits A to F, inclusive) present fairly the financial position of the
Institution at June 30, 1953 and the results of its operations for the year then ended,
in conformity with generally accepted accounting principles applied on a basis con-
sistent with that of the preceding year.

Haskins & Sells
August 31, 1953

XXV

Exhibit A Balance Sheet, June 30, 1953

Assets
Current Funds:

Cash in banks and on hand $594,117.64

Advances:

Departmental Research Operations 13,330.78

Other 423.73

Accounts receivable 469.01

Inventory of books 89,924.00

Deferred charges 23,250.49

Due from Endowment and Other Special Funds 343,939.44 $1,065,455.09

Endowment and Other Special Funds:

Cash in banks $286,954.59

Securities (valuation based on market quota-
tions at June 30, 1953— $50,224,304)—
Schedule 1:

Bonds $25,250,223.88

Preferred stocks 4,073,132.15

Common stocks 15,119,908.90 44,443,264.93 44,730,219.52

Plant Funds:

Investment in real estate and equipment — Exhibit F 5,049,163.44

Total $50,844,838.05

Liabilities

Current Funds:

Accounts payable $2,559.33

Reserve for valuation of books and accounts receivable 90,393.01

Current Funds Surplus — Exhibit C:

Appropriated unexpended balances $448,531.15

General Contingent Fund 492,321.45 940,852.60

Unexpended balance of restricted gifts and grants — Exhibit D 31,650.15 $1,065,455.09

Endowment and Other Special Funds:

Due to Current Funds $343,939.44

Principal of Funds — Exhibit E:

Capital funds $41,379,485.59

Special funds 3,006,794.49 44,386,280.08 44,730,219.52

Plant Funds:

Bequests, gifts, and income invested in plant $4,991,159.04

Harriman Fund — donated land 17,934.40

Harkavy Fund — donated land 2,070.00

Hale Fund— Solar Laboratory 38,000.00 5,049,163.44

Total $50,844,838.05

XXVI

Exhibit B Summary of Current Income and Expenditures

For the Year Ended June 30, 1953

Current Income:

Investment income:

Interest and dividends on securities $1,943,224.04

Less: Amortization of bond premiums 25,935.94 $1,917,288.10

Market value of stock dividend 13,137.00

Total— Schedule 1 $1,930,425.10

Less: Income added to Special Funds (Exhibit E) — Schedule 1 1,485.66

Remainder — Income available for current purposes $1,928,939.44

Other income:

Sales of publications $5,030.95

Dormitory and mess hall 9,125.01

Miscellaneous 605.00 14,760.96

Restricted gifts and grants utilized for current purposes — Exhibit D 60,932.75

Total current income $2,004,633.15

Current Expenditures (including expenditures against appropri-
ations of prior years) — Schedule 4:

Administration $231,245.79

Departmental Research Operations 1,312,527.64

General Publications 29,450.39

Research Projects, Fellowships, Grants, etc 79,590.13

Pension Fund — annuity and insurance 156,369.46

$1,809,183.41
Gifts and grants— Exhibit D 60,932.75

Total current expenditures 1,870,116.16

Excess of current income over current expenditures — Exhibit C $134,516.99*

*SUMMARIZED AS FOLLOWS:

Current investment income in excess of Trustees' authorized
appropriations during the fiscal year, credited to General
Reserve Fund— Exhibit E $87,250.44

Market value of stock dividend credited to Special Income

Reserve— Exhibit E 13,137.00

Revertment of unexpended current appropriations to General

Contingent Fund— Schedule 4 92,720.54

Reserved from this year's appropriations for current liabilities

and commitments — Schedule 4 124,087.49

Total $317,195.47

Less: Amount included in current expenditures, applicable to
allotments and unexpended balances from prior years'
appropriations— Schedule 4 182,678.48 $134,516.99

xxvn

Exhibit C Summary of Current Funds Surplus

For the Year Ended June 30, 1953

Balance, July 1, 1952 $902,323.05

Additions:

Excess of current income over current expenditures — Exhibit B . . $134,516.99
Transfer from Special Funds (General Reserve Fund) to cover
miscellaneous expenditures in connection with the building
program at Cold Spring Harbor— Exhibit E 4,400.00 138,916.99

Total $1,041,240.04

Deductions:

Transfer to Special Funds — Exhibit E:

General Reserve Fund, representing excess of investment in-
come over Trustees' appropriations for current purposes. . $87,250.44
Special Income Reserve, representing market value of stock

dividend received during year 13,137.00 100,387.44

Balance, June 30, 1953, per Schedule 3 $940,852.60

XXVlll

Exhibit D Summary of Current Restricted Gifts and Grants

For the Year Ended June 30, 1953

Unexpended Additions — Deductions — Unexpended

Balance Gifts and Grants Expenditures Balance

July 1, 1952 Received (Schedule 4) June 30, 1953
Departmental Research Operations:

Department of Plant Biology:

Research Corporation $1,125.47 $1,125.47

Department of Genetics:

American Cancer Society No. EG 2 IE 175.10 $7,400.00 $6,904.06 671.04

American Cancer Society No.

INSTR-72 4,000.00 4,000.00

U. S. Public Health Service No.

RG-149 C6 2,549.62 8,804.40 8,087.41 3,266.61

U. S. Public Health Service No.

E-133 C2 2,762.94 8,415.26 10,430.22 747.98

National Foundation for Infantile

Paralysis 600.00 600.00

Department of Embryology:

American Cancer Society No. CP 2F 206.50 206.50

U. S. Public Health Service No.

C550-C4 4,747.78 4,747.78

Research Projects, Fellowships, etc.:

Carnegie Corporation of New York:

Biology:

Point Lobos, California 4,245.32

Yerkes Laboratories of Primate

Biology 12,000.00

Geology :

Tilley, C. E 620.63

History:

Dantzig, Tobias 8,000.00

Physiology:

Russell, G. Oscar 1,522.44

Plant Biology:

Chlorella studies 11,979.02

Terrestrial Magnetism:

Magnetic polarization of the earth 4,428.42

Clews, C. J. Birkett 7,500.00

Miscellaneous:

Hale, George Ellery, biography . . . 1,000.00

National Academy of Sciences:

Tamiya, Hiroshi 500.00

12,000.00 .

620.63

4,000.00

300.00

1,222.44

9,500.00

2,479.02

1,887.88

2,540.54

1,368.90

6,131.10

1,000.00

500.00

Total $43,363.24 $49,219.66 $60,932.75 $31,650.15

XXIX

<
W

w
w

tn
Q

P
fa

w
fa

«
fa

ffl
H

O
Q
<
H
fa

w
U

fa
o

C/3

i— i

m
i— i

ffl
fa

T— (

i-H

i— i

re)

1— 1

>— >

q
of
CO
^5=

i— 1

•— »

cu >>>

4h Ph

««

« tH*

i— 1

!>■

i— 1

T»H

t>-

l>-

i— 1

*-«.

i—l

T— 1

cn"

<tf

T— 1

i-<

<L>

-3

4-3

-t->

•d J

3 '

-!->

1-H

■ • • €£= • •

CU ^

3 £

4->

a a

3 r

3 3

/—.

i—l

d c

CO
CN

, H

fa •

B 5

a s

5 >
d

^h"

.3
u
OT

■ o

I— i

• tH

• €©= ■

CU '

n -u

B ,

(Ll C

3 3

d fa

< n

x
x

i— i

o
o

•<*

On
O

"*
^

CO
<*

T— 1

CO
CN

T— (

co"

NO O

oq o

CN '. '. O

CN . . . LO

"*..... CN

t-T CN~

<tf

*>-

T— 1

t^.

i— 1

T-t

i-H

T— 1

i—l

J>-

*tf

TjH

*-.

J>»

*tf

J>-

-tf

i—l

1—1

1— i

i—i

C
3

-M

C
CD

T3
C

CU
>

^ «» b

d a3 o
W U U

1 §

cd cd

<U G
cu cd

H >

d
'o

(72

T3 co
3 Ph

.y cu

^5 -v

W 3
. fa

w >,

OX) 3
cu cd

3
fa

«4H

_cy

^ *2
S

cd fa

4-> CU
CU l-i

"E 5

cu c

co a

CU J 3

^% cu

5 °

° 3 S

d /P 3

f fa fa

cd cd <±3

'u "G ^

CU cu o

a a>

C/3 t/J t>

00
CN

no"
00
CO

• On

• 1— 1

. t^

t^.

. ^

*tf

. O

• 00

• OO

• 1—1

o
o

CO
CN

1—1

1— 1

!>•

NO*

co'

1—1

i-H

l-H
4&

00
CO

no"

i-H

^5=

1— 1

°°L
co"

On
CN.

CO"

"<*

fa
<

H
O

m co bo

8 3 .5

« ~ "2

£ 2 cd

Hd +->

•-- CO

«, c3 *»

£ M 3

5 cd O

b 1 °

O t4 d

^3 1* .a

-5 CN s

•s o\ o

Cd tH <l>

cd co

«2 a 5

h H 1)

S >. d

u bo co

cu c o

fa 'C '*->

-a 3 S

CO nj
co S
(U «

O
X

" a
v a

.3 cd

d
cu

-rH CU

in -d
o +*

"2 g

d o

•5 °

-3 •»

5 §2

cd t*J

tH IT)

*" On

CO 1H

cd -

* s

4_> cu

cd cu

d M

o 5 -

a o
bo d

d ^

3
j2

+->
M

S
o
cu

d
cd

■a

cu
cd to

°. d

2 3

3 ^
^ +->

■* cd
& u

«l

cd pj
co m

d M

3 d

fa 's

- a

d w

_cd 13

fa "3
O^

co bO
P d

O 55

r-* lH

cd ^

cu CU

-d x

« 2
co d

2*^

cu JH d y

> CU .2 -h

O A 4J (J

o fa cd 3

O u

fa "S
* &

o g

s °

3 O

XXX

W
O

w
u

fa
o

C/5

H
PQ

9 «

rt 2

)-i Q

<u -a

a 2.

° w

>.
u

T— 1

w J
r5

^ M

O -?

>—>

cd ^

i— j

o S

n a

H-) <J

M "8

M g

3 3

CQ ^3

^ 3
3

t*5

0) s

o "

d O

•a ?

pq g

o»

»-»

«u

'+3

•a

C/3

K> O

d v
3 -»-»

.2 o

<J CO

., *D

CU CN
CU in

3 ^

pq Xt

t^.

""*

r-^

no*

i— i

CN*

oo"

no'

i— i

°1

""*

CN^

i—i

oo~

©*"

t>T

©~

oo"

"<*

no"

NO*"

S0=

On
On

CO
4^

T-i

•• bX)

W o

a«

« ^

ta .. .
O

o u

.S|S

o s.

S a)

^ OX)

»- c

n °

ox) cu
O co

OX)
crj

bx)

C/J •— I

ex ^

S 6

2? crj

C 'S

o u

en o

d U
o

S c

+-> crj

d w

3 crj

O Ph

t—

!>.

i>."

T— 1

LO*

t^."

no'

LO*

!>.

t^.

i—l

1—1

T— |

T— 1

4£=

i— 1

1>"

t—

o S

+-> °

Si ^

a <

OX)

b c

o "^

0) o

E S

5 crj

€©=

NO O

t--

J>.

NO NO

oo'

CO*

i—

1—1

i — 1

i-l

i — i

l>-

1— i

oq.

t^.

i—

*3-

^J-

CO*

no'

'^ 1

rr-

T— 1

t^-

1—

•<*

i — 1

i— t

i— (

<k&

i-H

i— I

1>*

CO*

■*

On_

^©=

co"

1—1

LO"

i—i

'^h

i—

J>*

LO*

NO*

t^.

j>-

°°

t^.

«*

co_

co"

i—i

no"

^±

"!*

cn"

^5=

i-H

rj-

«*.

CN*

On*

T-H

T—

co'

t-~

i—i

t^.

!>■

i— i

***

i —

i—

i-H

r<:

^t-

i-H

co"

crj
CU
CD
CU

+->

crj

Q
*cd
o
H

**'

NO
ON

CN*

i>T

!>■'
LO

cn"

CO
00

CO*

ON
CO

CO
CN

!>.

i—l

NO
CN

CO*
On
CN

On"
CN
CO

O
H

cn (J

i— I

< o

OX)

w 13

U crj

fa "^
O

CO
4^

nO_

CO"
NO

!>.

On"
CN
0CJ

NO*

°1
CO"

NO_

cn"

co'

On"
CO

LO"

0C3

no"

NO
NO

co"

no"

NO
<*

€©=

NO
NO

NO*

On"
00

o
o

LA,

NO
NO

c/s

lO
<N

•^ -5 j cd
"o = ** w

T3 i, c C

c; c a; — '

S .& o a w

Q 3 TJ d O

S? w 1 1

c » V 2 §

.5 cu cu c -^

3 3

Oh O

.3 Si °

cu tn G 3 ^:
(-, CU •« 0) h>

^ CU

O, Kl

X o
W U

Q S

XXXI

Schedule 1 Endowment and Other Special Funds Investments as of

June 30, 1953 and Income Therefrom during the Year

Per Cent of
Total Investments

, * N

Book Market Income

Book Value Market Value Value Value Received
Bonds:

United States Government.... $10,809,443.35 $10,512,542 24.17% 20.81% $218,136.64

Foreign and International Bank 1,174,158.03 1,113,375 2.62% 2.21% 34,383.35

Public Utility 3,322,568.99 3,092,252 7.43% 6.12% 83,514.97

Communication 1,339,184.59 1,288,312 2.99% 2.55% 36,374.57

Railroad 641,530.52 651,855 1.43% 1.29% 26,137.69

Railroad Equipment Trust.... 1,075,956.31 1,022,468 2.41% 2.02% 23,708.16

Industrial and Miscellaneous.. 6,887,382.09 6,461,542 15.40% 12.80% 160,571.31

Total bonds $25,250,223.88 $24,142,346 56.45% 47.80%, $582,826.69*

Stocks :

Preferred $4,073,132.15 $3,714,826 9.11% 7.35% $167,379.70

Common 15,119,908.90 22,367,132 33.80%, 44.28% l,180,218.71f

Total stocks $19,193,041.05 $26,081,958 42.91% 51.63% $1,347,598.41

Cash $286,954.59 $286,954 .64% .57%

Total $44,730,219.52 $50,511,258 100.00% 100.00% $1,930,425.10$

* After deducting bond premium amortization of $25,935.94.

t Includes $13,137.00 representing market value of a stock dividend received.

% Income received allocated to Endowment and Other Special Funds as follows:

Funds, the income from which may be used for current general purposes — Exhibit B $1,928,939.44

Funds, the income from which is restricted to specific purposes — Exhibit E:

Bickel Fund $443.09

George E. Hale Relief Fund 162.87

Harkavy Fund 123.40

Woloff Fund 756.30 1,485.66

Total $1,930,425.10

XXXU

Schedule 2

Schedule of Securities

Principal
amount

Description

Approximate
Ma- Book market

turity value value

United States Government Bonds

$220,000 U. S. of America Treasury Bills 7-30-53

550,000 U. S. of America Treasury Ctf. of Ind. 2s 8-15-53

1,500,000 U. S. of America Treasury Ctf. of Ind. 2 ]/ A s 2-15-54

1,550,000 U. S. of America Treasury Notes 2 yis 1953

1,700,000 U. S. of America Treasury 2s 1954-52

3,150,000 U. S. of America Treasury 2 J4s 1962-59

1,000,000 U. S. of America Treasury 2 Hs 1980-75

400,000 U. S. of America Treasury 3 J4s 1983-78

50,000 U. S. of America Savings Series "G" 2 }4s 1954

100,000 U. S. of America Savings Series "G" 2 J4s 1955

100,000 U. S. of America Savings Series "G" 2 l / 2 s 1956

100,000 U. S. of America Savings Series "G" 2 yis 1957

100,000 U. S. of America Savings Series "G" 2 Ks 1958

100,000 U. S. of America Savings Series "G" 2 Ks 1959

100,000 U. S. of America Savings Series "G" 2 yis 1960

$10,770,000 Total U. S. Government

$218,884.11

$219,709

550,000.00

550,110

1,502,343.75

1,499,062

1,548,026.10

1,550,000

1,700,000.00

1,688,843

3,199,095.64*

2,978,718

991,093.75

948,750

400,000.00

398,750

50,000.00

49,600

50,000.00

49,300

100,000.00

98,200

100,000.00

97,600

100,000.00

97,000

100,000.00

96,400

100,000.00

95,800

100,000.00

94,700

$10,809,443.35

$10,512,542

$300,000
100,000
50,000
100,000
125,000
125,000
150,000
200,000

$1,150,000

Foreign and International Bank Bonds

Aluminum Company of Canada, Ltd., S. F. Deb. 3 Hs

Australia, Commonwealth of, S. F. 3 yis

Canadian National Ry. Co., 4>2S Guar

International Bank for Reconstruction and Development, 3s. .
International Bank for Reconstruction and Development, 3 ¥&s

Noranda Mines Ltd., S. F. Deb. 4 ^s

Shawinigan Water & Power Co., 1st Mtg. & Coll. Tr. S. F. 3s
Series "M"

Total Foreign and International Bank

1970

$304,256.16*

$304,500

1956

100,000.00

98,250

1957

48,750.00

48,500

1957

112,000.00

104,625

1976

125,000.00

114,375

1975

123,125.00

119,375

1968

154,546.87

150,000

1971

206,480.00*

173,750

$1,174,158.03

$1,113,375

Public Utility Bonds

$200,000 American Gas & Electric Co., Serial Note 2 yis 1954-55

125,000 Columbia Gas System, Inc., Deb. 3s 1975

237,000 Columbus & Southern Ohio Electric Co., 1st Mtg. 3 Xs 1970

300,000 Consolidated Natural Gas Co., Deb. 2 Hs 1968

150,000 Detroit Edison Co., Gen. & Ref. Mtg. 3 pis 1988

200,000 Minnesota Power & Light Co., 1st Mtg. 3 yis 1975

200,000 Northern Natural Gas Co., Serial Deb. 2 Hs 1954-55

100,000 Ohio Power Co., 1st Mtg. 3}{s 1968

200,000 Pacific Gas and Electric Co., 1st & Ref. Mtg. 4s 1984

200,000 Panhandle Eastern Pipe Line Co., Serial Deb. 2 Hs 1961-62

100,000 Panhandle Eastern Pipe Line Co., S. F. Deb. 3Xs 1973

50,000 Philadelphia Electric Co., 1st & Ref. Mtg. 2 V&s 1978

207,000 Philadelphia Electric Power Co., 1st Mtg. 2 ^s 1975

200,000 Public Service Co. of Indiana, Inc., 1st Mtg. 3 yis 1975

45,000 Puget Sound Power & Light Co., 1st Mtg. 4><s 1972

210,000 Tennessee Gas & Transmission Co., 1st Mtg. Pipe Line 2Ks. • 1966

191,000 Tennessee Gas Transmission Co., 1st Mtg. Pipe Line 3s 1969

93,000 Tennessee Gas Transmission Co., S. F. Deb. 4^s 1971

265,000 United Gas Corp., 1st Mtg. & Coll. Tr. 2 #s 1967

$3,273,000 Total Public Utility

$201,949.46*

$197,320

128,002.01*

114,375

247,970.16*

220,410

300,647.56*

286,500

152,250.00

153,750

203,658.13*

182,000

201,891.24*

198,210

101,500.00

100,625

203,560.00

205,750

201,625.43*

183,600

101,447.02*

94,000

49,687.50

45,000

210,846.72*

178,020

203,700.32*

184,000

46,209.73*

46,800

212,047.50*

191,100

195,640.42*

171,900

94,935.79*

95,092

265,000.00

243,800

$3,322,568.99

$3,092,252

* After deduction for amortization of premiums on bonds purchased subsequent to January 1, 1940.

XXX111

Schedule of Securities — Continued

Principal Ma-
amount Description turity

Communication Bonds

$150,000 American Telephone & Telegraph Co., Conv. Deb. 2 Ms 1961

150,000 American Telephone & Telegraph Co., Deb. 2 Ms 1975

200,000 American Telephone & Telegraph Co., Conv. Deb. 3 Ms 1963

200,000 Mountain States Telephone & Telegraph Co., Deb. 3 Ms 1978

100,000 New York Telephone Co., Ref. Mtg. 3 Ms 1978

200,000 Pacific Telephone & Telegraph Co., Deb. 3 Ms 1978

300,000 Southwestern Bell Telephone Co., Deb. 3 Ms 1983

$1,300,000 Total Communication

Railroad Bonds

$100,000 Chesapeake & Ohio Ry. Co., Gen. Mtg. 4 Ms 1992

284,000 Fort Worth & Denver Rwy. Co., 1st Mtg. 4 Ms 1982

100,000 Pennsylvania R. R. Co., Cons. Mtg. 4 Ms 1960

143,000 Western Maryland Rwy. Co., 1st Mtg. 4 Ms 1976

$627,000 Total Railroad

Railroad Equipment Trust Bonds

$150,000 Chesapeake & Ohio Ry. Co., Eq. Tr. 2s Guar 1956-58

300,000 Chicago Burlington & Quincy R. R. Co., Eq. Tr. 2 Ms Guar... 1958-63

100,000 Great Northern Railway Co., Eq. Tr. 2s Guar 1960-61

150,000 Pennsylvania R. R. Co., Eq. Tr. 2 Ms Guar 1958-62

100,000 Southern Pacific Co., Eq. Tr. 2 Ms Guar 1956 & 59

150,000 Southern Pacific Co., Eq. Tr. 2 Ms Guar 1956-58

150,000 Southern Railway Co., Eq. Tr. 2 Ms Guar 1956-58

$1,100,000 Total Railroad Equipment Trust

Industrial and Miscellaneous Bonds

$200,000 Allied Chemical and Dye Corp., Deb. 3 Ms 1978

200,000 Aluminum Company of America, S. F. 3 Ms 1964

187,000 American Tobacco Co., Deb. 3s 1969

234,000 Bristol Myers Co., Deb. 3s 1968

300,000 C. I. T. Financial Corp., Deb. 2 Ms 1959

150,000 Dow Chemical Co., Deb. 2.35s 1961

130,000 Dow Chemical Co., Conv. Sub. Deb. 3s 1982

153,000 Food Machinery Corp., S. F. Deb. 2 Ms 1962

275,000 Goodrich (B. F.) Company, 1st Mtg. 2 Ms 1965

255,000 P. Lorillard Co., Deb. 3s 1963

295,000 National Dairy Products Corp., Deb. 2 Ms 1970

488,000 Phillips Petroleum Co., S. F. Deb. 2 Ms 1964

200,000 Phillips Petroleum Co., Conv. S. F. Deb. 3.70s 1983

125,000 Pittsburgh Plate Glass Co., S. F. Deb. 3s 1967

150,000 Quaker Oats Co., Deb. 2 Ms 1964

300,000 Seagram (Joseph E.) & Sons, Inc., Deb. 2 Ms 1966

300,000 Service Pipe Line Co., S. F. Deb. 3.20s 1982

500,000 Shell Union Oil Corp., Deb. 2 Ms 1971

200,000 Sinclair Oil Corp., Conv. Sub. Deb. 3 Ms 1983

500,000 Socony- Vacuum Oil Co., Inc., Deb. 2 Ms 1976

100,000 Standard Oil Co., (Ind.) Conv. Deb. 3 Ms 1982

300,000 Swift & Co., Deb. 2 Ms 1972

500,000 Texas Corporation, Deb. 3s 1965

346,000 Union Oil Company of California, Deb. 2 Ms 1970

57,000 Union Oil Company of California, Conv. Sub. Deb. 3 Ms 1972

400,000 Westinghouse Electric Corporation, Deb. 2 Ms 1971

$6,845,000 Total Industrial and Miscellaneous

$25,065,000 Bonds — Funds Invested

* After deduction for amortization of premiums on bonds purchased subsequent to January 1, 1940.

xxxiv

Approximate

Book

market

value

$153,113.91*

$164,062

151,968.75*

135,000

220,853.78*

230,250

201,400.00*

194,000

101,183.45*

94,500

204,414.70*

190,000

306,250.00*

280,500

$1,339,184.59

$1,288,312

$99,464.29

$112,000

286,749.97*

285,420

104,662.50

105,000

150,653.76*

149,435

$641,530.52

$651,855

$146,340.34

$142,290

292,507.12

274,660

98,538.91

89,725

146,358.96

137,523

100,031.19*

93,730

146,251.10

142,565

145,928.69

141,975

$1,075,956.31

$1,022,468

$198,000.00

$200,000

200,000.00

199,750

189,236.11*

181,623

234,869.74*

225,810

300,000.00*

283,500

150,511.37*

141,750

131,922.84*

128,537

152,308.98

140,760

275,796.09*

262,625

260,387.35*

247,350

298,485.24*

271,400

492,889.73*

463,600

206,750.00

209,250

125,000.00

121,250

148,922.50

139,500

298,500.00

279,000

300,000.00

288,000

503,666.21*

452,500

192,416.25

193,750

489,528.75

437,500

101,286.72*

100,875

301,351.57*

267,000

519,645.49*

498,750

355,313.95*

318,320

57,000.00

57,142

403,593.20*

352,000

$6,887,382.09

$6,461,542

$25,250,223.88

$24,142,346

Number of
shares

Schedule of Securities — Continued

Description

Book
value

Approximate

market

value

400
1,000
1,500
2,000
1,500

500

600
1,900
1,125
1,000

600
2,075
1,000
2,000
1,500
1,000
1,000

300

800
2,000
1,300
1,500
1,000

900
2,000

760
1,300
3,100

35,660

2,000
5,300

19,360
2,600

10,250
2,850
3,000
4,700
2,800
7,500
5,200
2,000
60
1,500
5,400
2,000
2,750
1,250
6,373

12,200
1,000
3,500
6,100
4,800
7,980
2,500
3,620

Preferred Stocks

Air Reduction Company, Inc., 4.50% Cum. Conv. Pref $41,195.04

Anchor Hocking Glass Corp., $4.00 Cum. Pref 1 12,750.00

Appalachian Electric Power Co., 4K% Cum. Pref 159,000.00

Armstrong Cork Co., $3.75 Cum. Pref 205,500.00

Bethlehem Steel Corp., 7% Cum. Pref 183,637.50

Case (J. I.) Co., 7% Cum. Pref 62,225.00

Cleveland Electric Illuminating Co., $4.50 Cum. Pref 68,112.25

Consolidated Edison Co. of N. Y., Inc., $5.00 Cum. Pref 202,815.50

Continental Can Co., Inc., $3.75 Cum. Pref 115,312.50

Continental Can Co., Inc., $4.25 Cum. Conv. 2nd Pref 106,779.99

Corn Products Refining Co., 7% Cum. Pref 110,335.18

duPont (E. I.) de Nemours & Co., $4.50 Cum. Pref 235,401.89

El Paso Natural Gas Co., 4.10% Cum. Pref 111,442.21

General Foods Corp., $3.50 Cum. Pref 201,000.00

General Motors Corp., $5.00 Cum. Pref 187,937.50

General Shoe Corporation, $3.50 Cum. Pref 102,250.00

Grant (W. T.) Co., 3 H% Cum. Pref 100,447.91

Merck & Co., Inc., $4.00 Conv. Cum. 2nd Pref 31,200.00

National Distillers Products Corp., \ l A% Cum. Conv. Pref 80,000.00

Niagara Mohawk Power Corp., 3.60% Cum. Pref 207,990.00

Ohio Power Co., 4K% Cum. Pref 144,630.02

Pacific Telephone and Telegraph Co., 6% Cum. Pref 235,220.75

Panhandle Eastern Pipe Line Co., 4% Cum. Pref 104,166.68

Pillsbury Mills, Inc., $4.00 Cum. Pref 96,949.80

Reynolds (R. J.) Tobacco Co., 3.60% Cum. Pref 199,683.75

Sherwin-Williams Co., 4% Cum. Pref 83,764.83

Standard Oil Co. of Ohio, 3 H% Cum. Pref. "A" 139,976.28

U. S. Steel Corp., 7% Cum. Pref 443,407.57

Total Preferred Stocks $4,073,132.15

Common Stocks

Allied Chemical & Dye Corp $96,175.97

Aluminium Limited 221,852.22

American Can Company 445,351.97

American Cyanamid Co 161,924.48

American Gas and Electric Company 214,294.48

American Telephone & Telegraph Co 413,951.54

Anderson, Clayton & Co 164,768.84

Armstrong Cork Company 231,516.80

Atchison, Topeka and Santa Fe Rwy. Co 232,758.69

C. I. T. Financial Corporation 151,369.19

Caterpillar Tractor Co 272,536.55

Chase National Bank of the City of New York 71,361.04

Christiana Securities Co 356,143.00

Chrysler Corporation 68,005.61

Cleveland Electric Illuminating Company 227,773.54

Consolidated Edison Company of N. Y., Inc 67,530.37

Consumers Power Co 94,687.50

Continental Illinois National Bank & Trust Co. of Chicago 88,175.00

Continental Insurance Co 241,723.37

Continental Oil Co. of Delaware 239,598.64

Corning Glass Works 59,631.83

Deere & Company 117, 181. 35

Delaware Power & Light Company 128,803.87

duPont (E. I.) de Nemours & Co 195,905.06

Eastman Kodak Co 209,388.59

Fidelity-Phenix Fire Insurance Co. of New York 171,032.77

Fireman's Fund Insurance Co 128,858.78

$41,200

94,250

148,500

180,000

205,500

62,500

61,500

202,350

98,437

122,000

98,550

229,287

90,000

178,000

174,937

85,000

86,750

28,275

71,200

161,750

128,050

195,000

94,500

86,400

169,000

75,240

123,500

423,150

$3,714,826

$135,000
251,087
672,760
115,700
292,125
437,831
104,625
250,275
259,350
196,875
273,000

90,500
379,500
107,625
270,000

75,750
101,750
103,437
452,483
674,050

76,250

91,000
147,925
455,400
334,162
177,500
193,670

{Concluded on following page)
XXXV

Number of
shares

9,000
4,000
6,000
3,000
2,000

14,352

10,625
1,800
2,567

11,300
2,500
2,800
7,940
3,470.5
5,000
1,800
2,000
9,000
5,000
7,000
320
8,000

11,925
7,000
5,500
3,000
1,300
2,300

10,000
8,000
3,300
7,000
1,000
4,000
2,400
3,990
7,800
4,200
5,200
2,600

11,000
500
4,800
9,900
5,000
7,800
6,200
6,200
5,500
800
1,800
8,050
4,000
4,750
4,000
4,700
8,500
1,700

433,782.5

469,442.5

Schedule of Securities — Continued

Approximate

Book market

Description value v^lue

Common Stocks — Continued

General Electric Co $354,865.31 $639,000

General Foods Corporation 167,302.84 213,000

General Motors Corporation 159,935.38 359,250

Goodrich (B. F.) Company 155,957.20 200,250

Guaranty Trust Co. of N. Y 98,003.91 126,000

Gulf Oil Corp 310,532.07 640,458

Gulf States Utilities Co 223,614.04 233,750

Halliburton Oil Well Cementing Co 79,847.34 97,650

Hartford Fire Insurance Co 173,120.24 386,333

Humble Oil & Refining Co •. 188,600.61 700,600

Illinois Power Co 97,697.35 95,000

Inland Steel Company 156,896.54 109,550

Insurance Company of North America 243,482.20 647,110

International Business Machines Corp 213,004.45 798,215

International Nickel Co. of Canada, Ltd 185,533.15 208,750

International Paper Company 95,379.27 91,800

Johns-Manville Corp 74,008.06 120,500

Kennecott Copper Corporation 459,524.19 577,125

Kimberly-Clark Corporation J 227,813.88 200,000

Kresge (S. S.) Company 220,699.69 235,375

Mellon National Bank and Trust Company 67,193.07 100,800

Mercantile Stores Company, Inc 174,200.45 140,000

Merck & Co., Inc 223,709.20 232,537

Middle South Utilities, Inc 168,660.65 177,625

Minneapolis-Honeywell Regulator Co 127,075.73 313,500

Monsanto Chemical Co 126,552.58 258,000

Montgomery Ward & Co., Inc 70,154.45 76,700

National City Bank of New York 88,539.38 112,700

Newberry (J. J.) Co 132,451.57 320,000

North American Co 155,852.66 158,000

Ohio Edison Co 105,150.00 122,100

Penney (J. C.) Co 248,413.74 479,500

Peoples Gas Light and Coke Company 106,350.00 130,500

Pfizer (Chas.) & Co., Inc 82,503.08 112,500

Phelps Dodge Corporation -. 71,057.69 76,500

Philip Morris & Co., Ltd., Inc 210,697.55 199,500

Pittsburgh Plate Glass Co ' 251,672.40 378,300

Procter & Gamble Co 177,227.28 251,475

Scott Paper Co 137,909.13 308,100

Seaboard Oil Co. of Delaware 229,104.22 213,200

Sears, Roebuck & Co 251,140.45 646,250

Security-First National Bank of Los Angeles 35,039.88 50,000

Sherwin-Williams Co 294,227.78 321,600

Socony- Vacuum Oil Co., Inc 300,464.13 346,500

Southern California Edison Company 175,996.33 171,250

Standard Oil Co. of Indiana 282,824.78 560,625

Standard Oil Co. of New Jersey : 197,233.38 442,525

Texas Company 163,636.60 328,600

Union Carbide & Carbon Corp 172,416.63 353,375

Union Electric Co. of Missouri 16,787.72 17,500

United Fruit Company 51,075.58 95,400

United Gas Corp 127,987.59 193,200

United States Gypsum Co 362,416.20 426,000

United States Plywood Corporation 174,129.76 119,937

United States Steel Corporation 188,577.48 154,000

Virginia Electric and Power Co 92,038.13 109,862

Westinghouse Electric Corp 232,369.41 376,125

Weyerhaeuser Timber Company 88,953.47 93,500

Total Common Stocks $15,119,908.90 $22,367,132

Common and Preferred Stocks — Funds Invested $19,193,041.05 $26,081,958

Aggregate Investments (Bonds and Stocks) $44,443,264.93 $50,224,304

XXXVI

Schedule of Securities — Concluded

Summary of Security Transactions July 1, 1952 to June 30, 1953
Cash awaiting investment — July 1, 1952 $122,175.65

Bonds

Preferred Stocks

Common Stocks

Sale of Stock Rights.

Net Gain — To Exhibit E.

Sales and Redemptions

Gain

Loss

Book Value

$5,521.16

1,395,428.68

29,134.82

$1,430,084.66

$25,461.51 $3,142,425.89

165,458.90

1,588,283.18

$25,461.51
1,404,623.15

$1,430,084.66 $1,430,084.66

Income applied to amortization of bond premiums.
Market value of stock dividend

Total

Less: Cash transferred for current needs.

14,896,167.97
1,404,623.15

6,300,791.12

25,935.94
13,137.00

1,462,039.71
200,000.00

Remainder .

),262,039.71

A cquisitions

Bonds $5,441,286.65

Common Stocks 533,798.47

Cash awaiting investment — June 30, 1953

5,975,085.12
$286,954.59

XXXV 11

« 3

OiOOONOO'OOO
On ON>o'o6 4 s * "vf* 00

i— l On
tH

49^

On lo co tJh 00

!>■ On t-» co i— i

On On I s - t^ CO

Hifl^OCN
co th co i— i r-

CO
LO

oo"

"<*

1— 1

T— 1

CM
CO^

CN
LO

oq_

©'

OJ*

O
ir i.

cn"

i— 1
CO

cn"

On
49=

w
p

Q
W

cn
o
t/3

«JJ

i— i

>— i

Q lo

£ ^

< *- 1

NTS
30,

C/5
SZ5

o 5

u P

O I—,

< Q

cn a

p B

J £

fa w

P fa

C/3 «<

,„ W

1 b

Ph H

H fa

S? O

S ta

h- 1

ffl

\»

c/j

"5 g c

CD .t| C (7

5 fa w

5 ii

a o

CD ril c/}

5 2 ® S rS

"5 r a r Xi

£ fe w fe x

H fa)

o
CO

CJ Jlj

PQ .r?

C/3

-o

3
cu

1- ,

On On
CN O

"* o\"

co !>•

t— 00

oo i-T

1— I 1— I
49^4=9^

00 CN *^l i— I LO
On OO CN t-^ NO

CO^^OO iH CN
NI^VON of

cn t-h

t^iOOOOOOOcOLOO
t-H LO CN J>-' TjH rJH t^' 00*

OCSr- (NO-^T— I CN CO

t^cO'— HO^OONi^

*>■ "^ f— I CO LO l>Tr^o"

ONih^OnvOCNco
4p^'<-H CN Ol CO H

o
o

4©=

do
o o

"-lO.

OOcn
O VO cn
O* On oi

ONO
CN^CO__LO

CO ■<— i m£

OOOOOOOo

o o o o o o O o

dNoodio'dio

OloOOOcnOon
lO vO^O^O^sO^O^O OJ
CO NO t— ( NO CN ,— I ©"on"
x— I J>» t— I O OO tH Jl|

H i— ( CN x— l oi CO ,— i

CNOnloOnOnno-^v-)
OOOOx-hOOntHONcO
LO 1>" lo no' O* On NO* lo

i— iNOcO-t^ ,— !Nrt«
On no Tf On_0_co_CN oj

oCo'x-To x* co of lo"
€©=th oq hcn

NO
J>^

OI
LO

T— I

<3^

o
o

On
^9=

co^
of

o
o

J>.

€#

CO
0CJ

i>r

co '—I
co J>-

NO' NO*
O NO
NO LO

ON
J>.

LO*

*— I
CO
CN

On co no
co h -rfn

d d on

lo On no

■"^H LO CO

oToCno"

CN t^ LO

o
o

LO o

On O

O lo

CO O
O co

x* co
On lo

OnJ>"
00 T-i
O^VO

CN ^H

x* ON
LO CN

LO* co"

On 00

co'nO*"
CO

o o o o o
o o o o o

lo d d o" d
x>- o o o o

OO O^O^O lo
0\~lo lo"o"j>-"

O NO CO LO x*

CN th

lo co th On 00
NO *— i CN co NO

NO tjh J>" !>.' On
On lo lo 00 rH
VOi>^00^'* NO

ofcxfoft^Td"

CN lo On lo

NO
00

co'

o
<»

T— I

o
o

4©=

NO
ON

!>.

•>*

T— i

o
o

tJh"

o
o

Xt \

OO"
OI

c»

T— I

49=

LO
OI

49=

oo
d

ON
I

o
o

d
o

49=

NO
ON

"*~

▼—I
4^

o
o

oo"

OI
00

T— I

49=

cn
C
O

Oj
Ih

6
o

'-M >,

P m

aj biO
U o

a o

"73

logy
;ory.
gnet

OP"V .

, o

C/!l

r W

search
Plant Bi
Genetics
id Mess
boratory
Archaeo
Observal
trial Ma
Embrvol

•> 0)

c
c

"cj

fa
p p

U p

1 ^

ed:

lental Re
tment of
tment of
mitory ai
ysical La
tment of
t Wilson
of Terres
tment of

a
Q

c
.2

eratioi
blicati
rejects
nd....

c
H

'-P

p fa
gc75

+->
O

aj O, 3(1h 3

+j H In h U J5 li - .J-,
U U Oh OhQ O Oh X Oh O

- cijcua>' —l cuci>J-;ci)O

dminis 1

eneral

esearcl:

snsion

oaQQ OQ^PQ

3
CD

an
a^ 1

P^Dh

"c3

NO
NO

On
"■=*

3
ccj

-o

3
aj

tJh

co'

co"

j>-

°-

NO_

co"

lo"

"tf

49=

1— 1

CO*
00

CO
CN

NO*

i—l
i—l

On"
O
c»

1—1

4^=

49=

o
o

-<*

'^rh

©•

<4H

#©=

.11

■>->

<v-

i= ii

*=t

•

•E

o
O

*c3

o
o

CN*
LO
LO

OO"
CN
00-
1—1
4#

<^
H
O

O &.J3

flj »H

+-> a

a
o

a
a
% «J

4-j CD

> C o
aj 3 C
fa U fa

H
H

XXXVlll

£ c*

o p

lO CS lO fO 6 OO CO OS

vo > © , *-r©"*^'trro r-i

4^1— I CO r-KN^H

o
o

CO"

4/^

^ © CO tJh
co t"» 00 Ov

co t^^jToo

00'
cn"

T— I

3 "S

5
w
cu

co co

c3 O

•e o. d ^

0000colOOn©nO©
vOONf0^tNM>00

lO cn" J>-" CO* CO* LO O On

nO 1— if— lO <0 CN CO CN

omooof^ On On

rH^O\tNrH^OOO

On nO OOCMOOih

4©=i-H tH CN CO TH

u <

CN
CN

1— I

40=

fOrot^O

CN CO *-" ©

CO NO co O
0_v0 O «0

CO 1—1 LO t^
CN 1— I CN i*
CN i-H

CO
OS

no"
CN
VC5

1— I
43=

O
CN
1^

CO
O

cn"

NO
ON

CO
00

4/^

w
w

,— H

•3

t-»LO000000cO L OO

NO-^-rtli-HCNOLOO

HiotNN -^ rj" rj" CO
OCNi-HNOTfi— I CN CO

NfClrHlO'OONf^'O
t>» T^ tH fo"~LO~J>Ti— < o

Oni>-i-hi— iOnnOCNco
4=^i-l CN CN CO 1-1

t~«*

43=

On On co ^O

lo" O O On
rJH lo ON NO

CN 1^1 LO CO

1—1 On On"nO
CO CN t^. lo
CN i-H

co'

On
O
GO_

1— 1
4/^

NOOOOOCMMOI

3 CO rfl ON 1— 1 NC

5 co

■^COONt^LO^O^

H ^H J>; CO 00 ^

H . °^

S-.

LO CO CN no' co' NO* 1— 1 C

5 lo' lo" © Tt" O

' lo"

COCNCNOi— 1 00 CN O

J 00 co lo "*-|H nC

5 OO

lOCOcoOnOnOLOc/

5 i-H LO ^H \Q C

) i-H

'OONCXJOnO'-

h On © On 00 NC

> <*

i-h rf t^i CN ^ *>■ C

5 OO 00 CN ,_, li"

) t^

4=3=

i LO

4->

»• O • •>* CN 00 nO Td

H H 00 •

CN i-H

• OO CO LO NO 1-

h a

rH O

\ LO CN* cn"~0 c

* CO t~^

! no

CN H

.Nfoa^o

\ oc

J>-

O CN

. t^- rjn cn co lt

3 NC

i-H

CO NO

■NO ^ O -^

H LT

«*

4/5=

T— (

s -^

4^=

• • 0C

)

• • vo

.' ! rr

CO*

•s

. . NC

. . c<-

"oj

• * CN

• • LT

■ 4/=

4<i=

cOOnO^nOco^^-OC

) C

■ • irf

1— 1

On^iOvO^O^^

1— 1

CO
_CU

i-h" no' On tJh" 00* lo CN rt

J>*

ONOOOOOOONro^CN

CV)

OCNNOOr- ion00_nC

CO t>» CO t^ON lo © "d

1^-

cn"

1 — 1

l>- CN LO NO H "•* O

4©=i— 1 1— 1 CN CN

1—1
1—1

-G

-t->

OX)

03
CU
co
cu

O
cu

h Operat
t Biology
:tics

3 g

CU P

'h "0

ni l-i

3 *03
+->
G
cu

■ s

"-G

G
03

C
cu

tment of Plan
tment of Gene
mitory and M
ysical Laboral
tment of Arch
t Wilson Obsei
tment of Tern
tment of Emb

ct5
Q
CU

"ell

■4-:
03
Jh

+J
CO

.2
+j

-
cu

-J- 1

4-5

G
03
co

t+H

JhJhJh^GJh,hJh1h

oj cu an) 5 nJ rt

o3 u,

CU cy

G co
cu cu

u, Q, O.Q o'O.A ft O.
o3 cu cu •— 1 cu O J~ cu cu

cu
u

_o
"to

NO
i-H

no'

oq_

1—1

4i?=

**.

On
00

4^=

ON
i-H

NO"
NO
Ctt

CO
4^

CO*

co^

cn"

NO*
ON

4^=

<
H
O

o ©

cn ©'

LO ©
LO ©

CO LO~

1 i

4^=

o ©

TH 00

O CN

"2 «>

d ^
4 4^

©O
CN © ©
I/) . .
On CN O

•H ^ ©

H - cnO k
cJ J>- CN

CU OO

P 4^

CU
CO *q

cu o.

CU Oh

co CO

>> G

^Q O

•a '5

cu.5

> u

J^ p
a, Ui

a ft

03 ft

cu o3

-O a;

3 a

CN
LO
LO

O*
NO

OO"
CN

1—1

4*5=

-0

^G
CJ

LO i-l ©

On O ©

©' LO LO*

CO CN ©

irToT
4^

O
Si
u,

biO

*o
U

G
o3

j; h

G ••

.2CQ

03 .„ cn

uSi C

ft IS 9

* -£

ftW S _

ft I ~ E S

< co^ g O

fS-3 a co g

W Jh O CU —

J ,„ cu cu
^ cu

i* o. w

■ fl £S

hGc^I

_o
.2

a
a

ct3 -h-

■2*3.2
c +j ^

J- •- Jh

^•S ft.-

ft 3 t^

5 c ^ <u

-•-p £ a

G C*h G

CU O HH 5

G nJ cu cu
t >

•"■ Jh

o 53

> to

cu cu

NO
ON

CN*

1— 1
cO_
co"

i-H

4.0=

XXXIX

REPORT OF THE PRESIDENT

OF THE

CARNEGIE INSTITUTION OF WASHINGTON

FOR THE YEAR ENDING JUNE 30, 1953

REPORT OF THE PRESIDENT

OF THE

CARNEGIE INSTITUTION OF WASHINGTON

This Institution has a magnificent op-
portunity in the future if national condi-
tions are such that it can continue to
prosper, for the fields of fundamental re-
search in which it is engaged are by no
means exhausted or running into a phase
of diminishing returns. They offer in fact
many fine new vistas.

In astronomy we are just beginning to
reap the benefits of exceedingly powerful
instrumentation — to explore the far reaches
of the universe, assign it a revised measure
of magnitude, and throw light perhaps on
its ultimate geometry and its apparent ex-
pansion. With spectroscopy of enhanced
discrimination, we approach with new ef-
fectiveness the problems of the composition
of the universe and its evolution. As radio
astronomy joins optical methods, we see
things that are not visible to the eye or
even to the photographic plate.

The genetics of lower organisms is lead-
ing us deep into a new understanding of
the earlier phases of life processes, yet far
from the beginnings of life itself; for even
in bacteria and viruses the complexities of
the mechanisms of inheritance have been
found to be far greater than might have
been anticipated. The use of tracer tech-
niques, the speed and versatility of chroma-
tography, the extension of microscopy, the
selection of a few unique individuals
among billions by the use of phages — all
open up the possibility of understanding
better many of the life processes in their
more fundamental forms: the basic func-
tioning of the genetic system, the synthesis
of amino acids and the proteins, the de-
velopment of the human egg and embryo,
and much more. The day is coming when

man will use lower organisms far more
than he does today to produce food, to
separate raw materials from the sea for
his use, to synthesize the complex organic
compounds necessary for his industry and
for his physical well-being. Basic to this
use is an understanding of the funda-
mental life processes, and here is where
the Institution's interest lies, rather than in
the application. There is evident oppor-
tunity for another fifty years of exacting
endeavor in elaborating a constantly wid-
ening set of concepts.

Nor are we by any means far along in
our understanding of the planet on which
we perform such extraordinary investiga-
tions — its sea of air above with its chemi-
cal, meteorological, and electrical phenom-
ena; the earth beneath with its rocks, its
continents floating on a viscous mass, its
mountain building, and its mysterious
magnetic core. Here too the advance of
instrumentation provides new tools, elec-
tronic in nature to a large extent but also
chemical and physical in the more classic
sense. We wish to know how the earth
evolved into its present state, how it is now
composed, how to account for the mag-
netic and electrical performance we wit-
ness, how our occupancy of it will affect
some of its key features in the future.
Out of this may come better communica-
tions, new sources of materials, other
things that man desires. But we are chiefly
interested because we are human and have
an inherent urge to understand ourselves
and our environment.

The Institution thus has a varied and
fascinating future within its present scien-
tific fields if it retains its virility and its

CARNEGIE INSTITUTION OF WASHINGTON

skills. But will it be able to proceed as the
scene about it alters? It has thus far pros-
pered, and it has grown in strength and
distinction.

It is sometimes maintained that the day
of the independent research institution is
over, that henceforth fundamental research
belongs to the universities with heavy gov-
ernmental subsidy, and that such an insti-
tution no longer occupies a unique posi-
tion in the national scene as it once did.
When this view is based on more than a
casual estimate of national trends, it usu-
ally implies two criticisms: first, that the
research institution is isolated and that the
interchange and cross-fertilization of a
university community is necessary for best
progress in these days when disciplines are
becoming more intimately interlinked; sec-
ond, that fundamental research and the
education of brilliant students should be
closely related and that the research insti-
tution is prone to become ingrowing and
stodgy in the absence of the continual im-
pact of young minds.

These are cogent points, and the research
institution that wishes to preserve its health
had better pay attention to them. The
problem of isolation is mitigated when re-
search departments of an institution are
located close to a university, as some of
ours are. But isolation is more a matter of
attitude than of geography. It is quite pos-
sible for two departments of a university,
in adjacent fields, to occupy the same
group of buildings for years without a
trace of genuine collaboration or mutual
inspiration; such isolation is in fact ex-
emplified all about us. On the other hand,
a research department far from universities
may become a national or even a world
intellectual center in its field, the place
where those intimate interchanges between
masters occur which are often the seeds of
progress. In biology Woods Hole has per-

formed this function admirably; and in
genetics Cold Spring Harbor is now a
mecca, as are Mounts Wilson and Palomar
in astronomy. Interchange, the antithesis
of isolation, has its burdens and its abuses;
too much of the wrong sort can be a
nuisance. But the department or institu-
tion which gives the appearance of being
sufficient unto itself, which neither gives
nor takes, which tries to crawl into a hole
where it will be undisturbed in pursuit of
its inclinations, is sick and in need of a
radical operation.

One should not make the error of re-
garding formal education as the only
means of contact with youth. Formal edu-
cation is not the business of the research
institution. But if a research institution
allows itself to become old in its ways, the
fault is its own. It can and must, if it is to
maintain its vigor, expose itself constantly
to the influence of young minds. Much de-
pends on how they are selected and how
they are treated. Contact with a group of
graduate students in a university has its
needling effect and its invigorating influ-
ence; it also has its chores and involves
duties to students of only average ability.
The research institution, if it will, can se-
lect the cream, not only in mental capacity
but in the subtle characteristic of being
able to fit well into a scientific community.
Once selected, its younger staff members
can be made genuine participants in its
programs and in its group arguments and
plans. Organization within a department,
discipline of a sort, grades and titles, are
of course necessary. But there should be
no caste when a group of scientists meet
for a scientific discussion, in twos or tens;
and there should be no discrimination on
account of age or reputation. The young-
ster, in a gathering of two or fifty, should
be given his opportunity on the floor by
common consent, provided only that he

REPORT OF THE PRESIDENT, 1953

has something to say and knows how to
say it briefly and to the point. And i£ he
does not know how, he will soon learn,
with encouragement and in the right at-
mosphere, provided he has the stuff.
Though the problem of inducing partici-
pation on the part of younger members is
not unique to the research institution, it
is especially important to it; for the influ-
ence of the presence of youth is hardly
automatic in the institution's case, and
the dangers of getting into grooves and
of slipping into scientific fixity are greater
in a community of scientists alone than
in more diversified aggregations.

These are the prime dangers of the in-
dependent research institution. But they
are more than offset by its advantages.
The advantages could be enumerated and
examined at length. But only two will be
treated, two central points which are often
misunderstood.

It has long been realized that one of the
most effective ways to ensure outstanding
achievements in fundamental research is to
locate the individual of genius and sup-
port him liberally as he pursues his own
way. This is indeed important, and An-
drew Carnegie realized it fully, as is shown
by the early records of the Carnegie Insti-
tution of Washington. But there is more
to it than the simple statement indicates,
and the independent research institution
is in a good position to refine and develop
the simple idea to advantage.

We need genius, of course. But there
are geniuses who are utterly lone workers,
who are lacking in generosity and unduly
jealous of their reputations, who will re-
tire into a cubicle and work with no man.
Great things have been accomplished by
such individuals in the past and will be
in the future. But much more has been
and will be accomplished by those who
have brilliant creative skills combined with

human attributes that make them always
welcome in a community of their peers.
Their full accomplishment may not ap-
pear in their own papers, for they will be
surrounded by devoted disciples, and their
creative contribution will be in men as
well as in direct results. The independent
research institution furnishes an ideal cli-
mate in which such individuals can func-
tion to best advantage. Within it the pro-
gram of each department, created by the
staff itself, never handed down from above,
is more than the sum of its parts. Each
senior member formulates his own pro-
gram, of course; but he does it in con-
cert with his fellows, so that a depart-
mental program emerges, interrelated, mu-
tually contributory — a broad program for
progress on which all can embark with
enthusiasm. A university department can
do this too; but it is more readily done
by a research institution, free from the dis-
tractions of teaching and the manifold set
of relations that exists in a faculty.

The other part of the problem of sup-
porting genius is to find the genius before
he has matured and made a reputation;
for otherwise, changing his environment at
best merely gives him a wider opportunity
and at worst is a form of piracy. There are
occasional leaders in science and learning
equipped with a sixth sense who can do
this well. Such a one was Daniel Coit
Gilman, who, putting full time into the
effort for a considerable period, assembled
at Hopkins a brilliant congeries of schol-
ars: Rowland, Welch, Osier, Gildersleeve,
Remsen, Martin. But such gifted leaders
are far more rare than the geniuses they
seek. In fact it is doubtful whether genius
is as rare an attribute as it is generally sup-
posed to be, and whether there are not far
more men who would blossom into great
creativeness if given the chance. One
means of attracting men to a university

CARNEGIE INSTITUTION OF WASHINGTON

faculty is the august committee of selec-
tion, which passes on nominations and
brings men of distinction in from the out-
side. This means is not an effective one
for finding the promising youth at the
beginning of accomplishment, for commit-
tees very seldom take risks. It can, more-
over, be deadening to the morale of the
youngsters struggling up inside the organi-
zation. The best way to pick young men
of promise is to have the selection done by
a man in the same field, subject to checks.
This is the way it is done in the Carnegie
Institution of Washington. The person
who makes the initial selection is the direc-
tor of the department. His checks are
of two sorts. First, his senior staff partici-
pate with him, suggesting and judging;
and one can be rather sure that if they wel-
come a youngster with practical unanimity
he will fit, even though to a casual lay
observer he might seem to be rather a
queer specimen of the race. The other
check is by the President and Trustees,
who, though they can hardly compare indi-
viduals in detail, can certainly ascertain
that standards are held high and adequate
diligence is employed in the search. The
best way to get a new young scientist of
great creative ability into a group is to have
a group that is made up to a considerable
extent of men of just that character, and
then allow them to select their novices,
with provision for rotation and further
selection — all under the leadership of a
scientist with judgment. Judgment in the
selection of men is, in fact, one of the
primary attributes of a successful director
of research. Again, all this can be done
in a university department. But the tightly
knit group in a research institution can
do it better.

There are many things which a univer-
sity can do better than a research institu-
tion. But this is a discussion of funda-

mental research. In this field the research
institution is paramount and is by no
means an obsolescent form of organiza-
tion. It can, if it will, carry on funda-
mental research in a most fruitful way,
better on the whole than can be done any-
where else. So there is a place for it in the
future if conditions are such as to allow it
to prosper.

The principal condition, though not the
only one, for ensuring the prosperity of re-
search institutions is financial. It is, more-
over, a matter of endowments rather than
of current funds for projects.

There have been important trends in
the financing of research in this country.
Since 1920, according to a recent report by
C. I. Campbell, the fraction of the total
national income devoted to research has
been multiplied by ten. But the fraction
devoted to fundamental research has in-
creased relatively little. The federal gov-
ernment has entered strongly into sup-
port of research, but its entrance has
brought a host of problems, and its sup-
port is largely confined to projects, some of
them in basic research. These projects al-
low an organization such as a university
to expand its operations. But they do not
create new centers of research, nor do they
allow old ones to proceed to a permanently
secure basis of larger effort. The Car-
negie Institution of Washington has not
participated in government contracts, ex-
cept in the case of a few projects where
there seemed to be a duty to do a tem-
porary piece of work. It has not felt that
enlargement on the basis of such contracts
would be entirely healthy, and it is strongly
inclined to retain its complete independ-
ence. Although the federal program has
been an aid to many a college or uni-
versity in time of difficulty, it has its
dangers, and participation should be cir-
cumspect.

REPORT OF THE PRESIDENT, 1953

The great foundations have turned away
from endowment grants, and the new
foundations that are formed do not enter
this field. There are many reasons for
this. One valid reason is that giving away
money in great sums transfers the respon-
sibility for its use to others, and a founda-
tion often prefers to do its own thinking,
beyond the mere selection of a few institu-
tions from among many to receive its
largess for the furtherance of their sev-
eral purposes. This change in the policy
of foundations, however, adds to the great
sums being spent on a project basis. And
the temporary project is not a sound way
in which to carry on fundamental studies
of depth or subtlety.

The project idea, introduced largely dur-
ing the war and as a necessity at that time,
is far better adapted to applied research
than to fundamental research. This is part
of the reason why fundamental research
has not been expanded to the extent that
it should be. The foundations here have
to some extent missed an opportunity. As
the government entered strongly into sci-
entific research, they moved out. If they
had moved into basic research, they might
have preserved a balance. There was
plenty of opportunity for them to exercise
their talents of review and analysis in seek-
ing areas of opportunity, for the scope of
fundamental research has expanded far
more than the means for its furtherance.
There was hence opportunity to create
new institutions, endowed and independ-
ent, where they were most likely to pros-
per. A bit of this has been done; and there
are exceptions to the comments above, of

course. Carnegie Corporation donated five
million dollars for endowment to the Car-
negie Institution of Washington when it
was sorely needed. But in general the
foundations have not tackled the problem
of extending fundamental scientific re-
search in this country, nor is there any
great indication that they will do so.

New independent research institutions
have been established by individual philan-
thropists — the day of great fortunes and of
great altruism accompanying them is not
over. Yet these have been largely directed
at specific objectives, often in the medical
field, and often of a semi-applied nature.
This is not because men of wealth lack
interest in search for the unknown or ap-
preciation of the value, spiritual or aes-
thetic if you will, of the search for under-
standing, unencumbered by more imme-
diate objectives. It is rather that such men
understand better the sufferings and needs
of humanity and are anxious to alleviate
them.

This may be a reflection of our cultural
immaturity. As we proceed, there may be
more among us, highly successful in af-
fairs, anxious to serve humanity, who will
wish to look at the stars, or delve into the
earth, or probe for the secret of life, not
because it will add to the comforts or re-
duce the hazards of existence, but because
it may render us a more dignified and un-
derstanding race with greater satisfaction
in living. If so, there will be more and
greater independent research institutions
devoted to the search for knowledge for
its own sake.

Research Activities

The reports of the Directors of the De- At the Mount Wilson and Palomar Ob-

partments of Research may be read in de- servatories, H. D. and H. W. Babcock,

tail in the Year Boo\. I shall summarize taking advantage of the greatly increased

only a few highlights here. resolving power of the new Mount Wilson

CARNEGIE INSTITUTION OF WASHINGTON

gratings, and of new photoelectric tube nebula, using cepheid variables as distance
techniques that have become available indicators. He then proceeded to derive
since the war, have recently made an en- the distances of various other nebulae and
tirely new approach to the problem of the groups of nebulae extending out to the
general magnetic field of the sun. Their extreme range of the ioo-inch telescope,
new equipment permits them to make by referring them to the distance of the
automatically twenty to thirty uniformly Andromeda nebula as a standard. These
spaced traces across the sun's surface, re- very fundamental studies extended the
cording with greatly increased accuracy the diameter of the measured universe many
value of the magnetic intensity and po- thousandfold and revolutionized our ideas
larity for each point covered. The sun's as to the extent of the universe,
surface has been scanned in this way on But like all first measurements made at
most of the clear days for the past year, the extreme range of available equipment,
The new observations confirm the exist- these early results were very rough and
ence of a general magnetic field, much were subject to certain assumptions as to
weaker than earlier measurements sug- the properties of the distance indicators
gested, and limited to the neighborhood used. One of the first programs, therefore,
of the poles. Stronger fields, more local- planned by the nebular department for the
ized and transient, are also occasionally new Hale telescope was a step-by-step re-
found in the spot-free areas at lower lati- study of the problem of the distance scale,
tudes. These changing fields seem to af- This included a repetition of all measure-
ford an explanation for the occasional ments with the higher precision made pos-
large and erratic values of the earlier re- sible by the Hale telescope and by more
suits, which by the very laboriousness of advanced techniques now available. It also
the measurements were limited to a few involved a re-examination of the proper-
isolated points on the sun's surface. ties of the distance indicators used, in the

The new techniques clearly open up for light of recent discoveries concerning the

study the whole subject of magnetic fields various stellar populations of the near-by

and their fluctuations with time over the nebulae such as the Andromeda. Though

sun's surface. They also indicate the de- final results for all the steps are not yet

sirability of continuing daily scans of the completed, the preliminary values obtained

sun's surface for some time, preferably for by Walter Baade and his collaborators in-

the next sunspot cycle or two, in order to dicate that a drastic revision of the dis-

find what connections, if any, exist be- tance of the Andromeda nebula and other

tween these magnetic fields and other solar distances dependent on it will be required,

phenomena. Thus these tentative results lead to the

Very significant contributions have also conclusion that the earlier values of the
been made during the current year to an- distances and linear dimensions of all
other major problem that has been under nebulae must be approximately doubled
attack at the Observatories for the past and of their luminosities quadrupled. Fur-
quarter century. This is the problem of thermore, the old values of the nebular
the scale of the observable universe. masses, as determined from the linear di-

Edwin P. Hubble in the late 1920's sue- mensions and the relative radial velocities

ceeded in making the first definite meas- of the nucleus and periphery of the nebu-

urement of the distance of the Andromeda lae, must be doubled. Moreover, if we

REPORT OF THE PRESIDENT, 1953

assume the reality of the expanding uni- salts, the solubility in water drops off al-

verse, the age of the universe since the most to zero at the critical temperature of

expansion started from a common origin water. If the temperature is raised a small

must be doubled. amount above the critical point, but the

During the past year a new and ex- pressure is also enormously increased, solu-

tremely sensitive light-measuring device in bility again becomes measurable; and in

the form of a photon counter has been some cases, the gases have been able to

developed by William A. Baum for the dissolve more than their own weight of

Hale telescope. The new device makes it inorganic salt. Most of the work with

possible to measure the light received from volatiles has been done with water, but

an object at the extreme limit that can be it has been found that the inclusion of

photographed with the 200-inch telescope, small proportions of other gases may pro-

This limit was found to be at magnitude duce profound effects on the solubilities of

23, which indicates that the Hale telescope solids in hot, highly compressed gases and

has already reached objects considerably gaseous mixtures.

fainter than had been expected when the Bacterial cells can be conceived of as

telescope was planned. chemical processing plants wherein the

The experimental difficulties that have raw material from the culture fluid is
hampered work at the Geophysical Lab- utilized to provide the substance and en-
oratory with rock-forming minerals mixed ergy needed for the synthesis of a variety
with volatile components, chiefly water, of products. As a first step, the raw mate-
have now been surmounted by improved rial is converted to a group of intermedi-
techniques devised by George W. Morey, ates which are then used for further syn-
O. Frank Tuttle, and H. S. Yoder. The thesis. Simultaneously, a part of the raw
studies of these men are giving us a better material is oxidized to provide energy,
understanding of the formation processes These operations are nicely balanced to
of granite, the most common rock on the provide the proper supply of each inter-
surface of the earth. They have given us mediate. Experiments carried out during
a new insight into the chemistry of the the past three years at the Department of
ubiquitous micas, found in all rock types — Terrestrial Magnetism have given some
igneous, metamorphic, and sedimentary, insight into the kinetic chemical aspects of
They are yielding information pertinent growing bacterial cells. By measuring the
to geological problems ranging from vol- incorporation of a variety of isotopically
canism to metamorphism and the second- labeled compounds it has been possible to
ary enrichment of ore deposits. They are determine the rates of flow in many of
also opening up a new field of study in the reaction sequences of the cell. These
inorganic chemistry. Apparatus is now in flow patterns account for more than 80
current use which can give pressures up per cent of all the carbon incorporated,
to 200,000 pounds per square inch, at which is converted from glucose to the
temperatures of 1000 C or higher, and amino acids, purines, and pyrimidines sub-
studies with such apparatus have empha- sequently used as building blocks for the
sized that the solubility of solid inorganic synthesis of the giant molecules of protein
substances in compressed gases can reach and nucleic acid.

unexpected magnitudes. With silica and The flow in the Krebs cycle was studied

most minerals and with many inorganic with particular attention. This cycle has

CARNEGIE INSTITUTION OF WASHINGTON

long been recognized in mammalian tissue, the wind motions which produce them are
where its chief function is oxidation. The initiated by radiations and perhaps by par-
present studies show that in Escherichia tides from the sun.

coli growing with glucose as the sole car- The biochemical group of the Depart-
bon source, the Krebs cycle acts chiefly to ment of Plant Biology has been investigat-
synthesize aspartic acid and glutamic acid, ing the nature and function of the pig-
These amino acids are used in part directly ments by which the influences of light on
in protein synthesis and in part as inter- plants are mediated. As a key to the par-
mediates in the production of seven other ticipation of pigments in the photosyn-
amino acids, and thus supply more than thetic process, an effect of light on the
50 per cent of the carbon required for pigments during photosynthesis has often
protein. In this case the Krebs cycle is of been looked for. Such an effect was first
major importance in synthesis but con- discovered by L. N. M. Duysens, working
tributes little to oxidation. A similar situa- at Utrecht. He found reversible changes
tion has been found in an alga {Chlorella in the absorption spectrum of the particu-
pyrenoidosa), a. yeast (Torulopsis utilis), lar kind of chlorophyll that occurs in cer-
and a mold (Neurospora crassa). tain photosynthetic bacteria, which show

The flows in the Krebs cycle and in that the chlorophyll itself undergoes chemi-

other synthetic reactions are not rigidly cal changes induced by light. This year,

fixed. On the contrary, they are responsive as a Fellow of the Institution, he has found

to the conditioning of the culture medium, that these changes are increased when

Under certain circumstances the flow of oxidizing substances are added to the bac-

the Krebs cycle alters to emphasize oxida- teria and decreased in the presence of sub-

tion rather than synthesis. Two types of stances, such as peptone, that enter into

response to changing conditions have been the normal metabolism of the cells. This

studied in E. coli, an immediate change in shows a close connection between the cellu-

the flow rates due to a change in the chemi- lar metabolism and the pigments, and

cal environment, and a slower adaptive bears directly on one of the major prob-

change wherein the flow rates are further lems in photosynthesis, namely, how the

altered by changes in the enzymatic con- pigments are coupled to the enzyme sys-

stitution. This flexibility in the cell's re- terns.

sponse allows it to survive and prosper in a Leaves grown in the dark have a pig-
wide range of environmental conditions, ment, protochlorophyll, which changes to

New evidence has unexpectedly devel- the photosynthetic pigment, chlorophyll,
oped of the importance of atmospheric cir- by the absorption of light. James H. C.
culation and zonal winds in the produc- Smith and Allen Benitez have found that
tion of magnetic storms. The further elab- this change is remarkable in that it occurs
oration of the dynamo theory, made pos- in leaves far below their freezing point —
sible by modern studies of winds in the even at the temperature of —77° C. The
upper atmosphere coupled with recent rate of this change and its extent diminish
revisions in our understanding of the elec- as the temperature is lowered below the
trical conductivity of the ionosphere over- freezing point of water. The behavior at
head, now indicates that the electric cur- low temperatures suggests that two mole-
rents which produce magnetic storm effects cules of protochlorophyll, one of which is
are primarily in the atmosphere, and light-activated, interact to form chloro-

REPORT OF THE PRESIDENT, 1953

phyll. Glycerine extracts of protochloro- has been shown that this nucleic acid has
phyll from leaves have been prepared that the composition of the nucleic acid found
also show this light-activated transforma- in the infecting phage and that it is, in
tion. This glycerine-extracted protochloro- fact, a precursor of phage particles formed
phyll changes in the dark to another form later on. These findings are consistent
of protochlorophyll with a different color, with genetic results suggesting an intra-
This finding is significant because it has cellular population of multiplying non-
clarified the puzzling discrepancy between infective phage and with the idea that the
the previously measured action and fluores- noninfective phage particles consist of or
cence spectra of protochlorophyll in leaves, contain nucleic acid.

The influence of light on the consistency Studying the mechanism of genetic
of protoplasm in plant cells, which has changes, with maize as an experimental
been shown in water plants, has this year material, Barbara McClintock of the De-
been found by Hemming Virgin to occur partment of Genetics has found that the
in several land plants. Measurements of presence of a mutation-inducing factor at
the relative effectiveness of different colors a particular locus in a maize chromosome
of light on the phenomenon demonstrated has resulted in the production of a large
that only blue and blue-green light have number of mutations affecting the action
the effect of changing the viscosity of the of genes to either side of it. Some of the
cytoplasm of terrestrial plants. mutation-inducing events affect only the

One of the most important properties of action of an immediately adjacent gene,
living organisms is their capacity to repro- Others alter the action not only of this
duce; and one of the greatest puzzles of gene, but also of those located beyond it.
science is the question of how reproduction Thus a single mutation-inducing event,
is accomplished, in terms of chemical and occurring at one particular locus in a chro-
biological processes. A. D. Hershey, of the mosome, can result in the spread of muta-
Department of Genetics, has been studying tional change along the chromosome. In
reproduction in bacteriophages. His earlier the case studied, the spread included a re-
work suggested the possibility that infec- gion known to be at least six crossover
tion of the bacterium Escherichia coli by units in length. Spread beyond this limit
bacteriophage T2 starts off by injection of could not be determined because of the
pure phage nucleic acid into the cell. This inclusion of genes whose altered action
possibility has now been partially con- had an adverse effect on viability,
firmed in the following ways: (1) So far The work of the Department of Embry-
it does not appear that any labeled amino ology during the year has pursued its cen-
acids present in the parental phage parti- tral aim of working out the development
cles are transferred to offspring particles, of function in relation to the development
(2) Chemical analysis of isolated phage of bodily structure in the embryo and fetus,
does not reveal any basic protein (usually Among the noteworthy advances is a study
associated with nucleic acid in chromo- by S. R. M. Reynolds and associates at
somes), or any protein other than the Oxford of the action of the blood circula-
coat material, which is metabolically in- tion, especially of the lungs of the new-
active during phage growth. (3) Rapid born animal just before and after the time
synthesis of nucleic acid begins immedi- of its birth. Before respiration begins the
ately after infection of a bacterial cell. It lungs require only a small amount of cir-

CARNEGIE INSTITUTION OF WASHINGTON

culating blood, sufficient to care for nour- in the uterus, it gives concrete evidence

ishment of the developing but as yet un- that human single-ovum twinning actually

used living tissue. Immediately after the does occur in at least one of the several

first expansion of the lungs after breath- ways that have been postulated on theo-

ing begins, large amounts of blood are re- retical grounds — in this case by the devel-

quired to provide for the respiratory ex- opment of the fertilized egg cell into a

change of oxygen and carbon dioxide for single blastocyst (hollow spherical stage),

the whole body. The sudden filling of the within which the inner cell mass, usually

lung vessels with blood is associated with single, divided into two separate embry-

extreme changes in blood pressure in the onic areas.

heart chambers and great vessels, and sets During the past season the Department

up conditions which are critical for the of Archaeology began large-scale excava-

infant. Full understanding of these changes tions at the ancient city of Mayapan, which

is necessary not only for physiological is now the center of the Department's

study, but also for enabling obstetricians to activities. The materials and raw data

deal effectively with emergency circulatory emerging from this work seem to indicate

stresses in the newborn infant. that some profound changes occurred in

The intensive investigation of the physi- the local culture during the late stage of

ology of uterine muscle by Arpad Csapo Maya civilization represented by the ar-

has been continued with the collaboration chaeological remains at Mayapan. For ex-

of Beni Horvath and Brenda M. Schofield, ample, the numerous fragments of effigy

and from it much has been learned about incense burners of pottery, apparently

the action of the ovarian hormones. Little household idols, in the shrine room of an

has been known about the precise way in excavated dwelling suggest a decline in the

which these "chemical messengers" affect importance of the old priestly and public

their target organs. The work on uterine religion and a turning to private religious

muscle has taken a step forward by reveal- practices carried on in the home. The de-

ing that progesterone, one of the ovarian cline in importance of old religious forms

hormones, produces its effects on muscle is further suggested by what may have

tissue by altering the rate of access of been a late encroachment of dwellings

potassium ions to the contractile tissue. into the ceremonial areas of the city.

The collection of early human embryos These are interpretations that must be sup-
at the Department of Embryology received ported by other evidence before much
a notable addition from Chester H. Heu- weight is given to them, but it is worth
ser, Research Associate, namely the earliest noting that the changes in religious prac-
known human twin embryos, of about tices indicated by the archaeological evi-
seventeen days' development. Because the dence seem to be confirmed by the Span-
specimen is complete, consisting of the ish descriptions of Maya culture at the
two embryos and intact membranes in situ time of the Spanish Conquest.

Staff

It is always pleasant to call to your at- Wilson and Palomar Observatories, was

tention the honors that have come to direc- awarded the Ives Medal of the Optical

tors and staff members of the Institution. Society of America, received the degree of

Ira S. Bowen, Director of the Mount Doctor of Science honoris causa from

REPORT OF THE PRESIDENT, 1953

Princeton University, and was elected Cor-
responding Member of the Royal Society
of Sciences at Liege. By invitation Walter
Baade, staff member of the Observatories,
spent May and June 1953 at Leiden Uni-
versity and at Groningen University, Hol-
land. Dr. Baade was elected a member of
the American Philosophical Society; a
Foreign Member of the Royal Netherlands
Academy of Sciences; and President of
Commission 28 (Extragalactic Nebulae)
of the International Astronomical Union.
The National Science Foundation named
him as Principal Scientist at a Symposium
on Astrophysics held at the University of
Michigan in June and July 1953. Edwin P.
Hubble gave by invitation the Darwin Lec-
ture of the Royal Astronomical Society
and the Cormack Lecture of the Royal
Society of Edinburgh. Jesse Greenstein
was elected President of Commission 29
(Stellar Spectra) of the International Astro-
nomical Union and was appointed Chair-
man of a panel of consultants in astron-
omy, advisory to the National Science
Foundation. Harold D. Babcock, retired
staff member, was awarded the Bruce
Medal of the Astronomical Society of the
Pacific.

J. Frank Schairer, physical chemist at the
Geophysical Laboratory, was elected to the
National Academy of Sciences on April
29, 1953. The Volcanology Section of the
American Geophysical Union re-elected
Dr. Felix Chayes as its Secretary for a
three-year term.

At the Department of Terrestrial Mag-
netism, Merle A. Tuve, Director, was in-
vited to serve for a two-year term on the
Visiting Committee of the Department of
Geology of Massachusetts Institute of
Technology. On May 11, 1953, he received
the degree of Doctor of Science honoris
causa from the University of Alaska for
fundamental research contributions to stud-

ies of the upper atmosphere and for his
distinguished service over the years in con-
nection with the Geophysical Institute of
the University. Harry W. Wells, a staff
member of the Department, was made a
Fellow of the Institute of Radio Engineers
on November 6, 1952, "in recognition of
his contributions to ionospheric research
and the organization of a world-wide net-
work of ionospheric stations."

M. Demerec, Director of the Depart-
ment of Genetics, presented a paper at the
Symposium on Evolution organized by the
Society for Experimental Biology and the
Genetical Society of Great Britain, held at
Oxford, England, in July 1952. In the same
month Alfred D. Hershey, microbiologist
of the Department, participated in a Sym-
posium on Immunochemistry at the meet-
ing of the Society of Biological Chemists
in Paris.

George W. Corner, Director of the De-
partment of Embryology, spent the year
at Oxford, England, where he was George
Eastman Visiting Professor in the Uni-
versity of Oxford and Fellow of Balliol
College. While in England he gave the
Ciba Foundation Lecture in London on
December 4, 1952. During his absence he
was elected Vice President of the National
Academy of Sciences on April 29, 1953.
Robert K. Burns, staff member of the De-
partment, received the degree of Doctor
of Science honoris causa from his alma
mater, Bridgewater College, Bridgewater,
Virginia.

The Peabody Museum, Harvard Uni-
versity, appointed H. E. D. Pollock, Direc-
tor of the Department of Archaeology, a
Research Fellow in Middle American Ar-
chaeology. While attending a meeting of
the 30th International Congress of Ameri-
canists in Cambridge, England, in August
1952, J. Eric S. Thompson, staff member

CARNEGIE INSTITUTION OF WASHINGTON

of the Department, was elected President
of the Congress.

Elias A. Lowe, Research Associate of the
Institution, with the Institute for Ad-
vanced Study, was elected an Honorary
Fellow of Corpus Christi College of Ox-
ford University, England.

Two members of the scientific staff of
the Institution have retired.

Paul W. Merrill, who has been associ-
ated with the Mount Wilson Observatory
since January i, 1919, retired on Septem-
ber 1, 1952. He served as editor of the
Observatory publications from 1939 until
his retirement, and from 1949 until his re-
tirement he was a member of the Observa-
tory Committee of the Mount Wilson and
Palomar Observatories. Dr. Merrill's field
of research is stellar spectroscopy with spe-
cial emphasis on long-period variable stars.
His book Spectra of Long-Period Variables,
which appeared in 1940, is a classic in its
special field. An earlier work, The Nature
of Variable Stars, issued in 1938, is particu-
larly enlightening to the layman. Dr. Mer-
rill is one of the foremost authorities on
spectrum analysis. His outstanding con-
tributions in the field of astronomy were
recognized by his election to the National
Academy of Sciences in 1929 and by the
Academy's award to him of the Draper
Medal in 1946 "for his many important
contributions to astronomical physics, in
particular, those relating to his researches
in stellar spectroscopy." The same year
he was presented the Catherine Wolfe
Bruce Gold Medal by the Astronomical
Society of the Pacific. One of Dr. Merrill's
recent contributions was a report to the
National Academy of Sciences on the dis-
covery that the element technetium, origi-
nally found in nuclear experiments, exists
in stars of type S, although it has not been
found to exist naturally on the earth.

Dr. Merrill is continuing his researches
and writing.

E. Carleton MacDowell, staff member
of the Department of Genetics, retired on
October 1, 1952, after thirty-eight years of
service. His early studies of the effects of
alcohol on reproduction and sex ratio in
mice stimulated his interest in studies of
growth. In 1928 Dr. MacDowell began his
pioneer investigations of leukemia, and
since that time he has carried on an inten-
sive research program dealing with basic
problems of leukemia. His work in this
field in collaboration with associates at
Cold Spring Harbor and pathologists and
physiologists elsewhere established that leu-
kemia is a malignant growth comparable
to cancer, and that, like cancer, it can be
transmitted from one individual to an-
other by transplantation of affected cells.
Many lines of leukemia, characterized by
specific differences, were isolated. Cross-
ing experiments showed simple Mendelian
segregation of resistance and susceptibility,
indicating that one important recessive
gene is responsible for the resistance. It
was also found that the transmission of
resistance depends partly on the age of
either the nurse or the mother. One of
the latest achievements of MacDowell's
group was the isolation from the leukemic
cells of an X factor which apparently has
the effect of inducing resistance to certain
transplanted leukemias. Dr. MacDowell is
continuing his research on the mechanism
that induces resistance to line-I leukemia
with Dr. J. Victor, of Camp Detrick,
Maryland, under a grant to the Biological
Laboratory of the Long Island Biological
Association.

With the deepest regret I must record in
this year's review the deaths of three of
the Institution's retired staff members and
one member of the active staff.

George R. Wieland, internationally

REPORT OF THE PRESIDENT, 1953

known paleobotanist, who retired on Feb- urements of optical properties of grains

ruary i, 1933, died on January 18, 1953, in only 0.01 mm in size could be quickly and

his eighty-seventh year. The Institution conveniently carried out; his contribution

supported his studies on living and fossil during World War I to the development

cycads from 1903 until 1941. He became and manufacture of optical glass; and his

a Research Associate in 1907 and in 1917 special studies as Chairman of the Com-

was made an Associate in Paleontology, mittee on the Study of the Surface Fea-

Wieland was also a member of the Yale tures of the Moon, which he carried on

University faculty from 1906 until 1935. from 1924 until 1939 at the Mount Wilson

He discovered the petrified forest of cycads Observatory. His outstanding work in the

in the Black Hills of South Dakota and field of minerals was recognized by the

was the founder and donor of the Fossil Mineralogical Society of America when it

Cycad National Monument there. presented him with the Roebling Medal

George R. Wait, retired staff member on November 14, 1952.
of the Department of Terrestrial Magnet- The sudden death of Edwin P. Hubble,
ism, died on April 9, 1953, at the age of staff member of the Mount Wilson and
66. He played an important part in the Palomar Observatories, on September 28,
discovery of the world-wide simultaneous i953> cut short one of the most fruitful
pattern of daily variation in the earth's elec- careers in the field of astronomy. Hubble
trie field. His studies of atmospheric ioni- joined the staff at Mount Wilson on Sep-
zation and the origin and maintenance of tember 1, 1919, and served continuously
the earth's electric charge furnished strong except for a period during World War II
support to the theory that thunderstorms from 1942 to 1945, when he was Chief of
of the globe are responsible for the steady the Exterior Ballistics Branch, Army Ord-
negative electric charge of the earth with nance Ballistics Research Laboratory at
respect to the upper atmosphere and iono- Aberdeen Proving Ground, Maryland. His
sphere. early observations with the 100-inch tele-
Fred E. Wright, retired staff member of scope on Mount Wilson revolutionized the
the Geophysical Laboratory, died suddenly ideas of astronomers regarding the mag-
at his summer home in Canada on August nitude of the universe. Before 1920 the
25, 1953. Since his retirement on Novem- universe was conceived of as one great
ber 1, 1944, he had occupied an office in stellar system, the Milky Way; but Hub-
the Administration Building, where he ble's studies showed that our Milky Way
worked on a revision of his authoritative system is only one of some hundreds of
Methods of Petrographic-Microscopic Re- millions of such stellar systems extending
search: Their Accuracy and Range of Ap- out to the extreme limit of observation
plication, which the Institution first pub- through our largest telescopes. His subse-
lished in 191 1. For nearly half a century quent measurements of the radial velocities
Wright was a leader in the development of of extragalactic nebulae led to the signifi-
methods for the identification of minerals cant conclusion that they were receding
by the petrographic microscope. He joined from us and that those farthest away were
the staff of the Geophysical Laboratory on moving faster than those that were nearer.
January 1, 1906, and brought great distinc- These observations were the basis of the
tion to the Laboratory through his perfec- now widely held theory of the expanding
tion of techniques whereby accurate meas- universe. Hubble's discoveries emphasized

CARNEGIE INSTITUTION OF WASHINGTON

the need of greater telescopes and provided
one of the chief reasons for the construc-
tion of the 200-inch Hale telescope on
Palomar Mountain. He assisted in the de-
signing of this instrument and carried out
the first observations with it. He served on
the Observatory Committee from the start
of the joint operation of the Mount Wil-
son and Palomar Observatories in 1948,
and his counsel in the operation of the

Observatories was highly valued. Hub-
ble's scientific eminence was widely recog-
nized. He received honorary degrees from
several of the great universities both in
the United States and abroad; he was
awarded a number of gold medals for his
achievements in the field of astronomy,
and was elected to honorary membership
in some of the most distinguished learned
societies.

Finances

During the fiscal year ended June 30,
1953, income from investments, exclusive
of stock dividends, amounted to $1,917,-
288.10, an increase of $28,718.52 over the
previous year. The yield at book value of
securities held during the year was 4.36 per
cent, and at market value on June 30, 1953,
the yield was 3.79 per cent. Appropria-
tions authorized by the Board of Trustees
for the fiscal year amounted to $1,828,-
552.00, and additional funds transferred
from the General Contingent Fund during
the year by the Executive Committee
amounted to $84,500.00, making the total
amount of funds authorized during the
year $1,913,052.00.

The proportion of the Institution's port-
folio invested in common stocks at the
close of the year, at market, was slightly
lower than a year ago. Common stocks
held at June 30, 1953 represented 44.3 per
cent of the portfolio, bonds 47.8 per cent,
preferred stocks 7.3 per cent, and cash 0.6
per cent, as compared with 49.2 per cent in
common stocks, 42.8 per cent in bonds, 7.8
per cent in preferred stocks, and 0.2 per
cent in cash a year ago. The book value
of securities held at the close of the fiscal
year was $44,443,264.93, an increase of

$1,052,891.21 over the previous year; and
the market value of the securities was
$50,224,304.00 as compared with a market
value of $52,546,226.00 a year ago.

Of the Institution's income from invest-
ments, 60.9 per cent came from common
stocks, 30.4 per cent came from bonds, and
8.7 per cent from preferred stocks, an in-
crease of approximately 3 per cent in the
proportion of total income received from
bonds, and a corresponding decrease from
common and preferred stocks, over the
previous year.

It has been possible for the Institution
to carry on its operations during the fiscal
year within its available income without
curtailment of any part of its major pro-
gram. Also, some adjustments have been
possible where necessary to relieve eco-
nomic pressures. The Institution's re-
serves have increased modestly and re-
main in a healthy condition.

Statements showing the financial posi-
tion of the Institution at June 30, 1953, to-
gether with a certificate of Haskins &
Sells, Certified Public Accountants, appear
with the report of the Executive Com-
mittee.

Trustees

The Institution has suffered an uncom- tees: Frederic A. Delano, Homer L. Fer-
mon loss in the death of three of its Trus- guson, and Lewis H. Weed.

REPORT OF THE PRESIDENT, 1953

Frederic A. Delano, a Trustee of the 1953, in his eightieth year. He was elected

Institution for twenty-two years, died in to the Board on December 9, 1927, and

Washington on March 28, 1953, at the age resigned on September 8, 1952, owing to

of eighty-nine. He was elected to the ill health. A member of the Auditing

Board on December 9, 1927, and served Committee from 1931 to 1939 and again

later as member and Chairman of the from 1945 to 1952, and a member of the

Auditing Committee, Secretary of the Committee on Terrestrial Sciences from

Board, and member of the Executive Com- 1939 to 1952, he always took an intelligent

mittee, as well as member of the Commit- interest in the problems of the Institution,

tees on Terrestrial and Biological Sciences, and his resignation from the Board was

His resourceful counsel and his prevailing accepted with regret,

keen interest in the affairs of the Institu- Ferguson's name was familiar to ship-

tion were severely missed when he re- builders all over the world. After gradua-

signed from the Board on October 13, 1949. tion from the Naval Academy in 1892 he

A prominent and effective leader in the studied naval architecture at the University

movement for rational city, state, and re- of Glasgow. The Newport News Ship-

gional planning, Mr. Delano contributed building and Dry Dock Company, of

much to the redevelopment of Chicago and which he was first superintendent of hull

New York; and as Chairman of the Na- construction and later general manager,

tional Capital Park and Planning Commis- president, and chairman of the board, is

sion from 1926 to 1942 he was largely re- world famous for its merchant and naval

sponsible for such improvements in the vessels. Under Ferguson's direction the

Washington area as the revival of the company built fifteen of the Navy's big

L'Enfant plan, the Mount Vernon Me- aircraft carriers including the Hornet and

morial Highway, and the George Wash- Midway, fourteen battleships, and other

ington Memorial Parkway. Navy vessels.

He was valued as a public servant by Lewis H. Weed died on December 21,

five presidents — Taft, Wilson, Coolidge, 1952, in Reading, Pennsylvania, at the age

Roosevelt, and Truman — serving in the of sixty-six. He served on the Executive

widely diverse capacities of member of the Committee from the time of his election

Commission of Industrial Relations, mem- to the Board of Trustees on December 13,

ber of the Federal Reserve Board, Colonel 1935- He was also appointed Chairman of

in the Army in World War I as head of the Committee on Biological Sciences in

Army transportation in Paris, Chairman 1937; and on March 19, 1947, he became

of the League of Nations International Acting Secretary of the Board until his

Committee on the production of opium in election as Secretary on December 12, 1947.

Persia, and Chairman of the National Re- Dr. Weed was most generous in the time

sources Planning Board. he gave to Institution problems, and his

He will be remembered with deep af- sound judgment in their solution was al-

fection by all who knew him and par- ways highly valued. His death leaves a

ticularly by the staff in the Administra- gap which it will be difficult to fill,

tion Building of the Institution, where he Within the relatively short span of his

made his headquarters during the later life, Dr. Weed rendered distinguished

years of his active life. service in the administration of medical

Homer L. Ferguson died on March 14, research. After serving as professor of

CARNEGIE INSTITUTION OF WASHINGTON

anatomy at Johns Hopkins University, he
became Dean of the Johns Hopkins Medi-
cal Faculty and Director of its School of
Medicine. In 1939 he was appointed Chair-
man of the Division of Medical Sciences
of the National Research Council, and
during World War II he also served as Vice
Chairman of the Committee on Medical

Research of the Office of Scientific Re-
search and Development. He was espe-
cially interested in the development of
medical and other scientific libraries and
was mainly responsible for the establish-
ment of the Welch Library at Johns Hop-
kins.

Vannevar Bush

REPORTS OF DEPARTMENTAL ACTIVITIES
AND CO-OPERATIVE STUDIES

ASTRONOMY

Mount Wilson and Palomar Observatories

TERRESTRIAL SCIENCES

Geophysical Laboratory
Department of Terrestrial Magnetism

BIOLOGICAL SCIENCES

Department of Plant Biology

Department of Embryology

Department of Genetics

ARCHAEOLOGY

Department of Archaeology

REPORTS OF RESEARCH ASSOCIATES

MOUNT WILSON AND PALOMAR OBSERVATORIES

Operated by the Carnegie Institution of Washington
and the California Institute of Technology

Pasadena, California
IRA S. BOWEN, Director

Observatory Committee

Ira S. Bowen, Chairman Jesse L. Greenstein

Horace W. Babcock Edwin P. Hubble

Robert F. Bacher Ernest C. Watson

The basic parts of the 200-inch Hale indicators; (2) an investigation by San-
telescope were completed late in 1949, and dage of the absolute magnitudes of the
most of the auxiliary equipment such as distance indicators by means of color-mag-
spectrographs and photometers was fin- nitude studies in globular clusters; and
ished in the succeeding two and a half (3) a redetermination by Baum of the
years. As rapidly as equipment became stellar photometric standards used in all
available, the various phases of the obser- magnitude measurements. Though the
vational program were started, and most exact final results of all these steps are
of these programs have therefore been in still not available, the preliminary results
progress for from 12 to 42 months. Be- detailed later in this report are such as to
cause of the time required to accumulate indicate that a drastic revision in the dis-
the necessary groups of plates or other tance scale will be necessary; that is, that
observations and to measure and interpret the distances of all objects outside our own
them, however, very few final results had Milky Way system must be increased by
been obtained prior to the current year. approximately a factor of 2. The revision

In general these observational projects applies, however, to the "unit of distance"
have been parts of the two broad programs only and does not affect the relative dis-
of the Observatories as outlined on pages tances now assigned to the various nebulae.
7 to 9 of the annual report for 1 948-1 949 This change in the distance scale is of
(Year Book No. 48). A very substantial very fundamental importance to all cos-
portion of the observing time of the Hale mological problems, since it not only in-
telescope during moonless nights has been creases the distance of all extragalactic
devoted to removing the outstanding un- objects by a factor of 2, but also increases
certainties in the distance scale of the extra- their linear dimensions by the same factor,
galactic nebulae. This has involved a It also doubles the mass of these nebulae
thorough re-examination of each step of and increases their luminosity by a factor
the process and has included: (1) the of 4. Furthermore, if we assume that the
identification of the stellar types used as observed red shifts of the spectra of these
distance indicators in the Andromeda neb- nebulae represent true recessional veloci-
ula by Baade assisted by Miss Swope, and ties, and if we combine these velocities
in a few more distant nebulae by Hubble with the new values of the distances to
and Sandage, followed by a redetermi- determine the time that has elapsed since
nation of the apparent magnitudes of these the objects left a common origin, we now

CARNEGIE INSTITUTION OF WASHINGTON

arrive at a value of 3.5 or 4 billion years.
This is double the former value. If this
time is interpreted as the age of the uni-
verse, the new value removes much of the
discrepancy that existed between the for-
mer value and the age of the earth as
determined by radioactive measurements.

Though a small part of this revision in
the distance scale was brought about by the
increased accuracy of present photoelectric
measurements over the older photographic
methods, the major part of the change
arises from the elimination of large un-
certainties in the absolute magnitudes of
the stellar types used as distance indicators
in the earlier investigations. The earlier
studies were made possible by Hubble's
discovery, in the Andromeda and other
nebulae, of long-period cepheids whose ab-
solute magnitudes were known approxi-
mately from statistical studies of the proper
motions and radial velocities of a few such
stars in the galactic system. Although the
absolute magnitudes, or, more precisely,
the zero point of the period-luminosity
relation, of the cepheids were determined
thirty-five years ago and have been under
more or less constant scrutiny ever since,
no major revisions were clearly indicated
prior to the investigations discussed in this
report.

During World War II, Baade, using
newly developed red-sensitive plates, had
been able to resolve the nucleus of the
Andromeda nebula into stars for the first
time. These studies led Baade to the con-
cept that the great spiral nebulae such as
our own Milky Way system or the An-
dromeda nebula are made up of two types
or populations of stars. Population I is
found along the spiral arms and has as its
most conspicuous feature the blue giant
stars. Population I is also accompanied by
great clouds of gas and dust. The stars in
the nucleus of the nebulae and between the

spiral arms are of population II, which has
as its most outstanding feature the red
giants.

This new understanding of stellar types,
combined with the greater power of the
200-inch telescope, has made it possible to
follow a new procedure the first step of
which was a direct comparison of the ap-
parent magnitudes of many of the dwarf
stars in a near-by globular cluster with
the magnitudes of stars of the same type in
the immediate neighborhood of the sun.
Since the distances of these stars in our
neighborhood are known from trigono-
metric parallax measurements, this com-
parison gives an accurate value for the
distance of the globular cluster. In the
second step a similar comparison of the
magnitudes of the brightest stellar types in
the globular cluster, all of which belong
to population II, with the corresponding
types in the population II regions of the
Andromeda nebula yields a value for the
distance of Andromeda which is independ-
ent of the earlier very uncertain values of
the absolute magnitudes of the cepheids.

One important by-product of this study
by Arp, Baum, and Sandage of the prop-
erties of stars of the globular clusters has
been a detailed determination of the color-
magnitude relations in the stars of these
clusters which are of population II. This
has brought to light marked differences
between these relations for the two popu-
lation types. Likewise, the great light-
gathering power of the Hale telescope
combined with the speed of its spectro-
graphs has made it possible to begin for
the first time studies with moderate dis-
persion of the spectra of population II stars
as represented by stars of the clusters. Re-
sults already obtained suggest substantial
differences in the chemical composition
and other characteristics of the two popula-
tion types.

MOUNT WILSON AND PALOMAR OBSERVATORIES

OBSERVING CONDITIONS

The precipitation on Mount Wilson was 1953- During the same period observations

again below normal, with a total of 21.20 were made with the 60-inch telescope on

inches. Solar observations were made on 283 nights and with the 100-inch telescope

336 days between July 1, 1952 and June 30, on 310 nights.

SOLAR RESEARCH

Solar Photography

Solar observations were made by Cragg,
Hickox, Nicholson, and Richardson. The
numbers of photographs of various kinds
taken between July 1, 1952 and June 30,
1953 were as follows:

Direct photographs 662

Ha spectroheliograms of spot groups,

60-foot focus 692

Ha spectroheliograms, 18-foot focus. . 1,020

K2 spectroheliograms, 7-foot focus . . . 53,000

K2 spectroheliograms, 18-foot focus. . 1,017

K prominences, 18-foot focus 978

Solar Activity

The magnetic classification and study of
sunspots and related phenomena have been
continued by Nicholson and Cragg. Co-
operative programs have been carried out
with the U. S. Naval Observatory, the Ob-
servatory of Kodaikanal, the Meudon Ob-
servatory, the University of Michigan, and
the Central Radio Propagation Laboratory
of the National Bureau of Standards.

During the calendar year 1952, solar ob-
servations were made at Mount Wilson on
316 days, 17 of which were without spots.
The total number of spot groups observed
in 1952 was 219, which was 75 less than
in 1951. The southern hemisphere with 115
groups was slightly more active than the
northern with 104.

The sunspot curve for this cycle fits that
of the last cycle most closely with a time
displacement of a little less than ten and
one-half years, the activity in mid 1952
matching in phase that of early 1942.

Therefore the first spots of the next cycle
may be expected late in the summer or
early in the fall of 1953, with the minimum
coming in 1954.

The monthly means of the number of
spot groups observed daily for the past two
and one-half years are shown in table 1.

TABLE 1

Daily number

Month

January. .
February .
March ....

April

May

June

July

August. . .
September
October. . .
November .
December.

1951

1952

1953

4.0

2.8

2.4

4.5

2.0

0.5

4.2

2.0

1.2

5.7

2.2

2.3

7.0

2.2

1.2

5.5

3.8

1.9

5.4

3.5

4.9

5.1

6.4

2.7

4.3

2.0

5.2

1.6

3.4

3.1

Yearly average 5.0

2.8

Seventeen solar flares of intensity 1, five
of intensity 2, and none of intensity 3 were
recorded at Mount Wilson in 1337 hours of
observing. The average number of flares
recorded per 100 hours of observing was
1.6 in 1952, 2.3 in 1951, and 5.5 in 1950.

Sunspot Polarities

Magnetic polarities in each spot group
have, as far as possible, been observed at
least once. The classification of groups
observed between July 1, 1952 and June 30,

CARNEGIE INSTITUTION OF WASHINGTON

1953 is indicated in table 2. "Regular"
groups in the northern hemisphere are
those in which the preceding spot has S
(south-seeking) polarity and the following
spot N polarity; in the southern hemi-
sphere the polarities are reversed.

TABLE 2

Hemisphere

Regular

Irregu

lar

Unclassified

North

South

62
49

20
32

Whole sun . .

111

Solar Granules

The direct photographs of the sun taken
since the minimum of 1934 have been ex-
amined by Richardson to obtain plates
suitable for a study of the solar granula-
tion. These plates have been marked in
the plate catalogue so that they are avail-
able at once to anyone desiring images of
exceptional definition.

Microphotometer tracings were made on
plates taken throughout the present cycle,
which began about 1943. The average
width of a granule obtained by measuring
the interval from maximum to maximum
on 21 runs was 2V9 with a standard devia-
tion of ±i"o. The distribution in size
around the mean was nearly symmetrical.
If the granules are regarded as circular
disks, this value for their size is too small,
since the tracing will cut across them at
random instead of along a diameter.

No apparent difference could be seen in
the size or distribution of the granules in
a preliminary study of the tracings taken
on plates scattered through the cycle-
Direct measures on the size of the granules
made on a measuring machine, however,
gave an entirely different distribution in
size from that obtained from the tracings.
The frequency in size rises to a maximum

at about i'.'o and then falls abruptly. This
is about the theoretical resolution of the
4-inch aperture lens with which all our
direct photographs are taken, a fact which
indicates that smaller granules might be
found with larger resolution.

Solar Magnetic Fields

Magnetic studies of the sun are of value
not only in solar physics, but for the un-
derstanding of other magnetic stars. Much
attention has been given by H. W. and
H. D. Babcock to the improvement of the
photoelectric apparatus mentioned last year
for the detection and measurement of weak
magnetic fields of the sun's surface. Fields
as low as one-half gauss can now be meas-
ured quickly and with confidence. This
is equivalent to the measurement of Zee-
man displacements only 0.0004 as g reat as
the width of the Fraunhofer line itself. A
new self -synchronous automatic scanning
system has been added. Solar magneto-
grams have been made nearly every clear
day at the Hale Solar Laboratory, and de-
velopment testing of the magnetograph has
been carried out.

One specific result is the definite con-
firmation of the existence of a general mag-
netic field of the sun, detectable only in
high heliographic latitudes, above about
±65°. Persistent fields, of positive polarity
near the north heliographic pole and of
negative polarity near the south pole, have
been under observation for nearly a year.
These can be interpreted only as a prop-
erty of the sun as a whole. The intensity
and the extent in latitude of these polar
fields are variable; rather frequently the
intensity is found to be weaker near one
pole than near the other. The mean in-
tensity has been of the order of one gauss.
These fields arise in those regions from
which the coronal features known as polar
tufts are seen during eclipses.

MOUNT WILSON AND PALOMAR OBSERVATORIES

At lower heliographic latitudes little or
no evidence of a general field is found,
but instead more localized and variable
multipolar magnetic activity occurs, even
in the absence of visible surface features.
Occasional extensive weak fields are also
observed. A considerable body of data is
being accumulated in these magnetograms,
and significant correlations with solar ob-
servations of other types are being investi-
gated. It will evidently be of value to con-
tinue regular observations with the mag-
netograph throughout the solar cycle. The
design of a second installation, with im-
provements, is now in progress.

The Abundance of Beryllium

The solar abundance of beryllium has
been determined by Greenstein, in collab-
oration with Mr. E. Tandberg-Hanssen, a

Fulbright scholar, on plates taken by H. D.
Babcock at the Hale Solar Laboratory.
The solar abundance ratio Be/Li appears
to be about 8, whereas on the earth and
in meteorites it is about 1/6. This sur-
prising excess of Be cannot be easily ex-
plained. Thermonuclear disintegration of
Be and Li at the center of the sun is very
rapid. Only in the outer envelope of the
sun does the lifetime of Be exceed that of
Li appreciably. The presence of Be indi-
cates that the circulation between the cen-
ter and the solar atmosphere is very slow.
To account for the known depletion of Li
in the sun, while Be is unaffected, we must
assume that the circulation penetrates
down into zones at a temperature of
3,500,000° C, but no farther. This is a
rather special assumption, but otherwise
the Be abundance cannot be understood.

PLANETS AND SATELLITES

A fast-moving asteroid (1953EA) was
discovered by A. G. Wilson on a plate of
the National Geographic Society-Palomar
Observatory Sky Survey taken with the
48-inch schmidt camera on March 9, 1953.
Its orbit can be well determined from
photographs taken in March by Wilson
and Harrington with the 48-inch, in April
by Dr. L. E. Cunningham with the 60-inch
telescope, in May by Cunningham with the
100-inch, and in June by Sandage with the
200-inch telescope.

When discovered it was 6,000,000 miles
from the earth, but its closest possible ap-
proach is about 3,000,000 miles, one-fifth
that of Eros. When closest to the earth its
magnitude is about 13. Its period is very
nearly 4 years, and there should be no
difficulty in observing it again in 1957.
Unless its albedo is much less than that
of the moon, its diameter cannot be greater
than one-quarter of a mile.

Positions of Jupiter's faint satellites were

obtained by Nicholson with the 100-inch
telescope at the 1952 opposition of Jupiter.

The Atmosphere of Jupiter

On the morning of November 20, 1952,
the occultation of o Arietis (2 b 48 111 52 s ,
+ 14 53') by Jupiter was observed by
Baum and Dr. A. D. Code, using a photo-
electric photometer attached to a grating
spectrograph at the Cassegrain focus of the
60-inch telescope. This event afforded a
rare opportunity to determine the mean
molecular weight of the gases composing
the outer layers of Jupiter's atmosphere
simply by observing the rate at which light
from the occulted star was extinguished.
Two optical effects are involved: (1) the
spreading of the star's light due to differ-
ential refraction, and (2) ordinary expo-
nential extinction due to molecular scatter-
ing. It was shown from theoretical con-
siderations that differential refraction is
the dominant effect and that the rate at

CARNEGIE INSTITUTION OF WASHINGTON

which the star fades from view depends
only upon the geometry of the occultation
and upon the scale height in Jupiter's strat-
osphere. The scale height is in turn re-
lated to the temperature and mean molec-
ular weight.

Observationally, the problem was to ob-
tain a light-curve for the fading of a Arietis
without serious interference from the thou-
sandfold greater luminosity of Jupiter,
against whose limb the event occurred.
"Seeing" precluded the use of a sufficiently
small focal-plane diaphragm to isolate the
star from Jupiter's disk, but the difference
between their spectra provided the means
of discrimination needed. Since the B5
spectrum of o Arietis has practically no
K line of Ca 11, whereas the reflected G2
(solar) spectrum of Jupiter has a deep
one, monochromatic photometry at that
wave length strongly favored Arietis.

The best fit of the observed light-curve
to a family of theoretical light-curves yields
a value of 8 kilometers for Jupiter's scale
height, corresponding to a mean molecular
weight of 3.3. This is the first direct obser-
vational evidence that the atmosphere of
Jupiter consists largely of hydrogen and
helium.

Motion pictures of this occultation of
Arietis by Jupiter were taken at the same
time by Pettit and Richardson at the
coude focus of the 100-inch. The camera
was started operating automatically about
15 minutes before the predicted time of
occultation, taking exposures every 3.5 sec-
onds. The progress of the planet toward
the star was also followed visually in blue
light through the eyepiece with a power of
1500. The time determined visually was
io h 48 111 12 s UT, in good agreement with
that determined independently by other
observers. The predicted time for Mount
Wilson was io h 52.o m .

Some 200 images were obtained, cover-
ing the complete event. The most interest-
ing feature of the photographic record is
the remarkable variations in brightness of
the star about 50 seconds before its final
disappearance. Since a close examination
of the star and the planetary images indi-
cates a reasonably steady state of seeing,
these variations have been attributed to
turbulence in the atmosphere of Jupiter.
Marked fluctuations in brightness were
also observed about the same time on the
photoelectric light-curve made by Baum
and Code.

STELLAR SPECTROSCOPY

During the report year all major tele-
scopes except the 48-inch schmidt camera
were engaged in spectroscopic observations
for approximately two weeks each month
during the period when the moon inter-
feres with direct photography and photom-
etry. Over 560 spectrograms were taken
with the 60-inch telescope, 960 with the
100-inch, and 750 with the 200-inch Hale
telescope. These spectrograms were ob-
tained chiefly to provide information con-
cerning the chemical compositions, tem-
peratures, pressures, turbulence conditions,
magnetic fields, etc., present in the at-

mospheres of stars and in planetary neb-
ulae.

The great light-gathering power of the
Hale telescope combined with the great
optical speed of its coude spectrograph,
especially with the shorter-focus cameras,
has for the first time made it possible effec-
tively to extend these studies to certain
major new groups of objects. These in-
clude: stars of population II, the nearest
definite examples of which are found in
the globular clusters; the white dwarfs;
and the faint blue stars near the galactic
pole.

MOUNT WILSON AND PALOMAR OBSERVATORIES

Long-Period Variable Stars

Long-exposure spectrograms, chiefly
with the 200-inch telescope, have been
studied by Merrill and show many details
of the extraordinary behavior of the rich
bright-line spectrum of the long-period
variable x Cygni near minimum light. Of
60 bright lines whose intensities increase
markedly during the two months preced-
ing minimum, one-third are forbidden
lines of Fe 11, and two-thirds are unidenti-
fied. Differences in radial velocity between
various groups of lines and between plates
at various phases are small.

A survey by Merrill and Miss Lowen of
numerous high-dispersion spectrograms of
long-period variables has shown that the
gallium line A4172.05 is regularly present,
and that it changes during the light-cycle
from absorption to emission in such a way
as to indicate that the emission comes about
through fluorescence, the exciting line be-
ing A4033.07 of Mn 1.

Shell Stars

Spectroscopic observations of two in-
teresting shell stars, 48 Librae and HD
33232, begun many years ago, have been
completed by Merrill. The oscillation of 48
Librae, HD 142983, which began about
1934 with a remarkable outward thrust of
the atmosphere, has been clearly observed
from 1938 to 1952. The observed outward
motion slowly increased until in 1937 the
velocity was about 100 km/sec; it then de-
creased and in 1939 changed to inward
motion. Two cycles of about 8 years each
have now been observed, minimum veloci-
ties (in the algebraic sense) having oc-
curred about 1937 and 1945, maximum
velocities about 1941 and 1950. The ampli-
tude has diminished as if the oscillation
were damped. At times of minimum ve-
locity the radial velocities derived from
successive lines in the Balmer series of

hydrogen have a negative progression from
Ha to W35; at times of maximum velocity
the progression is positive. Thus the ultra-
violet Balmer lines toward the limit of the
series yield a larger amplitude than do the
lines nearest to Ha. Metallic lines, par-
ticularly those of Fe 11, yield a velocity
curve similar to that from H2S. The curves
for H28 and Fe 11 precede the curve for
H6 by variable intervals averaging per-
haps 100 days.

The variable velocity observed in HD
33232 is now believed to be due to a slow
atmospheric oscillation rather than to mo-
tion in a binary orbit as was previously
supposed. The cycle appears to have
lengthened from 3710 days to more than
4000 days, minima having been observed
about 1924 and 1948, a maximum in 1930.
The shape of the velocity curve also ap-
pears to have changed. The amplitude of
the velocity curves derived from metallic
lines and from hydrogen lines near the
limit of the Balmer series exceeds 100 km/
sec. This amplitude is greater than that
yielded by hydrogen lines near Hy, thus
following a pattern that is probably charac-
teristic of oscillations in shell stars.

Other shell stars, not so extensively ob-
served, that probably behave somewhat like
48 Librae and HD 33232 are (3 Monocerotis
A, HD 44351, HD 183656, and HD 220300.

Merrill and Miss Lowen have intercom-
pared 21 shell stars for which spectrograms
with dispersion 10 A/mm are available.
Relative intensities of lines of hydrogen
and of seven metals were estimated. Lines
of Mg 1, Ti 11, and Cr 11 have rather strong
direct correlations, whereas those of Sen
and Ni 11 show an inverse correlation with
each other, although with considerable scat-
ter. Atomic data offer a partial explana-
tion of these facts. In 9 stars the central
absorption of A4026 He 1 ranges from 3 to
12 per cent, the actual width from 5 to
12 A.

CARNEGIE INSTITUTION OF WASHINGTON

In most shells the velocity varies slowly
through a large range, and the structure of
the lines shows small changes during the
velocity cycle, wings appearing on the side
toward which the lines are moving. In
stable shells the lines are symmetrical and
sharp. Atoms that form a shell probably
drift slowly away from the photosphere,
but, through a slight outward acceleration
over a long interval, may in some stars at-
tain velocities as great as ioo km/sec. The
possible nature of the outward force was
briefly discussed.

SUBDWARFS AND WHITE DWARFS

White dwarfs have been somewhat neg-
lected in the past because of their faint-
ness. The coude spectrograph at Palomar
permits observation of these interesting ob-
jects at relatively high dispersion. Green-
stein is measuring the line profiles of the
hydrogen lines in the brighter white
dwarfs at 38 A/mm. The lines have wings
about 200 A wide; in some white dwarfs
they are relatively deep, in others so shal-
low that even Hy and Hh are invisible on
the plates but can be found on tracings as
great shallow depressions. A detailed study
of the yellow "white dwarf" Van Maanen
2 is being made; this has a lower tempera-
ture and shows a few metallic lines.

The known subdwarfs down to magni-
tude 1 1. 5 are also being observed by Green-
stein in the range of spectral types O to G8.
The general features of the later sub-
dwarfs are: weak lines of the metals, lack
of rotation or turbulence, and relatively
great strength of hydrogen. The frequency
of binaries seems to be small. It seems
probable that the subdwarfs will show a
wide range of spectral peculiarities, par-
ticularly in the weakness of the metallic
lines at a given temperature. Parallaxes are
badly needed to ascertain whether the
spectral peculiarities are correlated with
luminosity.

The early-type subdwarfs are extremely
interesting. A very great range of lumi-
nosity and spectral peculiarities exists
among the faint blue stars. Two new
helium-rich faint stars have been found;
in one, some of the He 1 lines are 9 A wide,
owing to pressure- and Stark-effect broad-
ening. One Bo star has no detectable hy-
drogen lines and strong helium, nitrogen,
and silicon.

Blue Stars near the Galactic Pole

A number of the brighter blue stars dis-
covered by Humason and Zwicky near the
galactic pole have been observed with the
200-inch coude by Greenstein and Munch.
The stars of spectral types B8 to Ao seem
to be normal and presumably are proto-
types of the blue stars on the horizontal
branch of the globular-cluster diagrams.
The earlier stars observed have spectra that
are somewhat peculiar, especially in the
absence of ionized carbon. It seems pos-
sible that some of the effects of age in these
stars of population type II are now being
observed spectroscopically; that is, the nu-
clear-energy processes have resulted in the
exhaustion of hydrogen, with formation
of helium and possibly of carbon. The lat-
ter will then be converted into nitrogen in
the envelope surrounding an exhausted
core. The early stars of population II so
far observed seem to have only small rota-
tion; hot stars in our neighborhood tend to
have large rotations.

Stars in Globular Clusters

In order to find out whether the blue
objects found by Humason and Zwicky at
the galactic pole belong to the population
of the halo of the galactic system, some of
the blue stars in globular clusters are being
observed for comparison by Munch. So
far only 5 stars in M 3 and M 13 have been
photographed with the Newtonian spectro-
graph of the 100-inch telescope. But it is

MOUNT WILSON AND PALOMAR OBSERVATORIES

planned to continue this program until a been found with only thermal motions in

fair-sized sample of spectra is obtained. its atmosphere. With high dispersion it is

To facilitate classification, widened coude found that as the luminosity increases in

spectra at 38 A/mm have been obtained, the G stars, the turbulence increases, the

by Deutsch with the Hale telescope, of a rare earths strengthen more rapidly than

number of red giants in the globular clus- other ions, and the CN strengthens. (The

ter M 92, and of a smaller number in M 3. latter effect is greatly reduced in stars of

The spectra in M 92 fully confirm Baum's the high-velocity group, which mimic stars

findings on the abnormal weakness of the one luminosity class lower.)

metallic lines and the strength of the In collaboration with Dr. L. H. Aller, a

Balmer lines. From estimates of a number quantitative analysis of 4 subdwarfs has

of metallic-line ratios that are sensitive to been initiated by Greenstein. The spectra,

temperature, however, it is found that the at 10 A/mm, are of HD 19445, 140283,

excitation temperatures of these stars are 161817, and 219617. Three of these are the

not greatly different from their color tern- so-called "intermediate white dwarfs" of

peratures. To the extent that their spectral Mount Wilson Contribution No. 511; HD

types indicate excitation temperatures, the 161 817 is a star of very high velocity,

types of these stars may therefore be said Coude spectra of most of the metallic-

to be in fair accord with their color classes, line stars in the Hyades and in the Coma

The material at hand, including many Berenices cluster have been examined by

spectrograms by O. C. Wilson and Bowen, Deutsch. The apparent abundance anom-

tends to support the thesis of Sandage and alies found by Greenstein from an analysis

Schwarzschild that the metals-to-hydrogen of the spectrum of t Ursae Majoris appear

ratio is appreciably lower in M 92 than in not to be a common feature of all these

the other globular clusters studied, and objects, and there is considerable doubt

that this ratio varies systematically from whether elements with second ionization

cluster to cluster in a way that is correlated potential in the neighborhood of 13.5 volts

with the position of the red-giant branch show a common behavior, as Greenstein

in the color-magnitude diagram. found to be the case in t Ursae Majoris.

In the open cluster NGC 663 measures

Chemical Abundances and Physical Con- have been made by Deutsch of the param-

ditions in Stellar Atmospheres eters characterizing the Balmer discontinu-

Greenstein's program for the spectropho- ity in the spectra of several of the most

tometric measurement and first analysis of luminous early-B supergiants. Though the

the composition of the normal G stars with measures are not precise, they suffice to

high dispersion has been completed. Dr. confirm the prediction of W. Becker, based

Jean Humblet, of the Institut d' Astro- on his heterochromatic photometry, that

physique, Liege, added two peculiar stars these objects exhibit very abnormally

to this program, 31 Aquilae (a so-called strong Balmer continua. The interpreta-

A4150 star) and HD 18474, a carbon-poor tion is obscure.

star. The turbulence of the G stars is All the observational material for the

found to be directly correlated with the eclipses of £ Aurigae in 1 939-1 940 and in

luminosity; turbulence is found also in the 1947— 1948 has been measured and analyzed

high-velocity giants. It is interesting to by O. C. Wilson and Abt. The only model

note that the subdwarf Groombridge 1830 which can be made to fit most of the

has zero turbulence; no star had previously observational data is one in which the

CARNEGIE INSTITUTION OF WASHINGTON

chromospheric matter occurs in condensa-
tions of some kind.

Spectrum Variables

Corrections have been found by Deutsch
for the previously published periods of two
spectrum variables. For the "strontium
star" x Serpentis the correct period is 1.60
days; the older value was the beat period
of this with one sidereal day. The new
observations of this star also reveal that
midway between the phases of Ca 11 maxi-
mum and Sru maximum there appear a
group of lines much sharper than all others
in the spectrum, and possibly attributable
to La 11. The correct period of the "silicon
star" 56 Arietis is 0.726 day. In this star,
while the silicon lines and the associated
unidentified lines go through a single oscil-
lation of intensity, the helium lines vary
through two full cycles.

An object that exhibits a spectrum simi-
lar to that of 56 Arietis is HD 124224,
which last year was found to have the
very short period of 0.521 day. Radial-
velocity measures made with the assistance
of Miss Burd, over the half-cycle centered
on silicon maximum, indicate that, whereas
the Balmer lines and Mgn 4481 yield a
constant velocity, Si 11, Fe 11, and the un-
identified lines that vary with these yield
velocities that change from —40 km/sec
to +40 km/sec.

The objects named above are 3 of the 9
spectrum variables whose periods have
been published. In most of these stars, the
lines are so wide that the Zeeman effect has
not been observed. In two of these, HD
125248 and a 2 Canum Venaticorum, H. W.
Babcock has found magnetic fields that
vary synchronously with the line strengths.
With Miss Gjellestad, he has also found
periodic field variations in HD 153882;
this object is not known to be a spectrum
variable. Since all these stars are members

of Morgan's sequence of peculiar A-type
stars, Deutsch has grouped them together
and has shown that the periods are closely
correlated with the line widths. Over the
whole range of periods from 0.5 to 20 days
the line widths on coude spectra vary
roughly in inverse proportion to the peri-
ods. In fact, it appears that the widening
of the relatively invariable lines in these
spectra can be quantitatively understood
on the assumption that each star rotates
with the period of variation of its spectrum
or of its magnetic field, and that we view
the star from a point nearly in its equa-
torial plane.

Accordingly, Deutsch has been led to the
hypothesis that the periodic magnetic and
spectrum variables are stars having their
magnetic axes highly inclined to their ro-
tational axes, and having in their atmos-
pheres large abundance anomalies which
depend primarily on magnetic latitude.
Additional evidence for this general pic-
ture seems to be found in the radial veloc-
ities of HD 124224 as well as those of 56
Arietis and a 2 Canum Venaticorum. The
observations of line doubling reported by
W. A. Hiltner and other observers and the
observations of the crossover effect in line
contours appear to be consistent with this
hypothesis. On the other hand, as is
pointed out in the next section, this simple
rotational hypothesis does not by itself
appear to be capable of explaining many of
the magnetic observations of these stars.
Attempts are being made by Deutsch to
give a theoretical account of such rotating
magnetic fields, and to discover whether,
by accretion of interstellar ions accelerated
to cosmic-ray energies, these fields could
produce the observed abundance anomalies.

Stellar Magnetic Fields

The observations of stellar magnetic
fields made in the past seven years have

MOUNT WILSON AND PALOMAR OBSERVATORIES jo

been summarized by H. W. Babcock. Among magnetic stars showing irregular

With the assistance of Miss Burd and Mrs. magnetic changes are HD 4174 and HD

Deutsch, fields have been measured in 35 133029. In the latter the fluctuations are

stars, of which 27 are peculiar objects of quite rapid.

types A and F. Six are spectrum variables. Two interesting magnetic variables that
All these stellar magnetic fields are variable may have an important bearing on the in-
to a greater or lesser extent: 11 have al- terpretation of these objects are HD 71866
ways shown a positive magnetic polarity, and HD 173650. The former has the most
10 negative, and 14 reverse polarity from pronounced crossover effect yet found, and
time to time. HD 173650 shows a very abnormal mag-

The magnetic changes are regular in netic behavior of chromium. The star has

some stars and irregular in others, with sharp lines, some of which develop faint

pronounced spectrum variation appearing emission wings at certain phases. These

in some but not all of the regular mag- stars are among those currently under ob-

netic variables. The irregular magnetic servation for the determination of period

changes can be interpreted only as the re- and amplitude.

suit of intrinsic variations in the stellar The magnetic curve of a 2 Canum Vena-
field, not of rotation of a star having a con- ticorum reported last year is difficult to
stant field. Attempts to devise a physical interpret because of the relatively sudden
model for the magnetic stars have not yet reversal, in about one-tenth of the period,
led to a wholly satisfactory result. Here it Preliminary tests at the crossover phases
is necessary to take into account not only failed to reveal plane polarization in the
the magnetic variations in detail, but also line profiles, to be expected if the transverse
the velocity and intensity effects shown by Zeeman effect predominates as in an ob-
various elements, as well as the crossover lique rotator. Spectrophotometric meas-
eflfect, line profiles, and variations in light ures of the lines of spectrum variables as
and color. a function of phase are lacking, but partial

For the long-period magnetic and spec- series of calibrated plates have been ob-

trum variable HD 188041, observations tained for HD 125248 and for a 2 Canum

have been collected for six years. Accurate Venaticorum with the aim of supplying

measurements of the field are possible be- such data,
cause of the number and sharpness of the

lines. The varying magnetic field is regu- Interstellar Lines

lar in period but irregular in amplitude; During the fall of 1952, about 50 early-

the changes cannot, therefore, be explained type stars in the northern Milky Way

by rotation alone. At magnetic minimum, were photographed by Munch with the

which recurs at intervals of 226 days, the coude spectrograph of the Hale telescope,

polar field is about + 1400 gauss. Between in the region of either the Ca 11 or the Na 1

minima, the field fluctuates rather irregu- lines. The study of these plates has re-

larly from cycle to cycle, ranging upward vealed that stars between galactic longi-

sometimes to +4800 gauss. The star is a tudes 65 ° and 135 and at distances (from

spectrum variable, in which lines of Eu 11, spectroscopic distance moduli) of the order

iSVn, and probably some other elements of 2 kiloparsecs show two well separated

vary with phase. The behavior of this star and about equally strong components. In

suggests some remote relation to the mag- comparison, near-by stars (up to distances

netic cycle of the sun. of 1 Kpc) show only a single strong com-

CARNEGIE INSTITUTION OF WASHINGTON

ponent. The velocities of the least dis-
placed components in distant stars, together
with the velocities of the interstellar lines
in the near-by ones, in the mean, define
quite well the galactic-rotation effect on
the radial velocity of a point at a distance
of about 500 psc. The shift of the most
displaced component in distant stars cor-
responds to a galactic-rotation effect of a
point at about 3 Kpc.
These observations have been interpreted

by Munch as indicating that the interstellar
gas clouds are concentrated along the struc-
tural features outlined by Morgan, Oster-
brock, and Sharpless as the spiral arms of
the galactic system. Further observations
are being carried out in order to extend the
range of galactic longitude where the large-
scale arrangement of the interstellar gas
clouds may be detected, and also to esti-
mate the degree of concentration of the
clouds along the spiral arms.

NEBULAR RESEARCH

Highlights in the year's work are repre-
sented by: (1) a definitive value for the
order of the unit of absolute distance in the
scale of extragalactic distances, (2) rapid
development of investigations of globular
clusters, and (3) successful trials of a new
"photon-counting" photometer which per-
mits accurate measures of luminosities to
the extreme limits of photography, and
probably beyond. The first item is the re-
sult of a sustained program of investiga-
tions which has now furnished a funda-
mental datum for mapping the sample of
the universe that can be observed. The
second item not only has established the
unit of distance, but is rapidly building up
our first picture of a pure type II stellar
population, and is contributing important
clues to stellar evolution. The third item
opens to investigation many quite new
fields near the threshold of photography,
of especial interest in problems of cosmol-
ogy and evolution.

RADIO SOURCES

As was mentioned in the last annual re-
port, Baade found a remarkable nebulosity,
apparently a chaotic mass of short fila-
ments, in the position of the strong radio
source Cassiopeia A. Minkowski has made
a spectroscopic study of the nebulosity and
finds that it represents a new type of object.

Individual filaments show large internal
motions; in one, about 2" wide and 20"
long, velocities range from —1000 to at
least +2200 km/sec. The spectrum of this
filament is unusual, showing lines of [O 1]
and [O in] about equally intense and at
least five times stronger than the unob-
served [On] lines and Ha. Other fila-
ments show smaller internal motions and
less anomalous spectra. The velocity differ-
ences between filaments are so small rela-
tively to the spread within filaments that
no decision is permitted as to whether the
nebulosity as a whole is expanding. If the
object is expanding, the velocity of expan-
sion is obviously small as compared with
the internal random motions; this condi-
tion is just the reverse of that found in
shells around novae and supernovae.
Therefore, it is not permissible to assume
that the nebulosity is a supernova shell.
Nor can it be such a shell decelerated by
interaction with interstellar gas. To con-
vert the velocity of expansion almost en-
tirely into random motions, an interstellar
cloud of outstanding density would be
needed, and evidently the nebulosity is not
inside a cloud of this type.

The radiation emitted by Cassiopeia A
has been studied by Dr. L. H. Aller and
Minkowski, who report that it appears to
be derived from the kinetic energy of free

MOUNT WILSON AND PALOMAR OBSERVATORIES

electrons. The spectrum may be under-
stood if the electron density lies between
io 3 and io 4 per cubic centimeter, and if the
electron temperature is less than 15,000° C.
Dissipation of mechanical energy in the
filaments may serve to maintain this elec-
tron temperature.

Minkowski and Greenstein have esti-
mated that the energy emitted per unit
volume by some of the strongest radio
sources is of the order of io -22 erg per
cubic centimeter. This rate is higher by
many orders of magnitude than that for
the galactic system. Compared with the
nonthermal radiation associated with sun-
spots, the radiation per unit volume of the
strongest sources is weaker by a factor of
the order of io 12 , but the radiation per
unit mass is about the same. The kinetic
energy available in the collision of gas
clouds at high velocities is large as com-
pared with the maximum radio emission
observed. Finally, the Cygnus radio source,
identified by Baade and Minkowski as a
pair of stellar systems in collision, emits
more energy in the radio-frequency spec-
trum than in the visual spectrum. Similar
collisions could be detected by radio tech-
niques at distances beyond the limit
reached by photography with the 200-inch
reflector.

STELLAR PHOTOMETRY
Photometric Standards in Selected Areas

Baum reports that the photoelectric
standardization of stars in ten Selected
Areas is now observationally complete to
magnitude 19. On the average there are
4 stars per magnitude interval in each area,
making a total list of about 400 stars. Re-
duction of the measures to magnitudes and
colors on the International System has been
under way for more than a year and is
now about three-fourths complete. These
magnitude sequences will be used as the

fundamental standards for Hubble's cur-
rent photographic work on clusters of neb-
ulae, for Baade's work on cepheids in the
Andromeda nebula, and for a variety of
other photographic investigations. The new
photon-counting photometer described un-
der "Instrumentation" will now be used
for providing extended sequences between
magnitude 19 and the plate limit in three
of the ten areas.

The Variable Stars h Ceti and UX
Ursae Majoris

Under the general heading of photom-
etry may also be mentioned photolectric
observations by Walker of two variable
stars. One, h Ceti, was followed on two
nights with the 20-inch reflector on Palo-
mar and was found to vary through a
range of about 0.025 mag. Though a
unique period could not be determined,
the data are consistent with the period of
0.16122 day found by Henroteau for the
variations in radial velocities. These facts
make it probable that the star is a member
of the (3 Canis Majoris class.

The second variable, the eclipsing star
UX Ursae Majoris, was observed in the
yellow, blue, and ultraviolet on seven
nights with the 60- and 100-inch reflectors.
The data confirm several peculiar features
of the light-curve and the short-period in-
trinsic fluctuations in brightness discovered
by Linnell, and contribute several new
features, including a bright shoulder on
the curve before the eclipse, in addition to
the previously known shoulder following
eclipse. The intrinsic fluctuations in bright-
ness are irregular in length, the interval
between successive maxima varying from
1 to 15 minutes. Amplitudes in the ultra-
violet (observed as large as 0.18 mag.) are
about three times those in the yellow.
These fluctuations together with the bright
shoulders on the light-curve suggest that

CARNEGIE INSTITUTION OF WASHINGTON

the attraction of the secondary produces a
bright, unstable spot on the surface of the
primary star beneath.

The depths of the minima are 1.02, 1.10,
and 1.22 mag., in the yellow, blue, and
ultraviolet, respectively, measured from
constant light. Although there is no phase
lag in the time of minimum light in the
different colors, discrepancies between ob-
served and predicted times of minimum as
large as 5 minutes have been recorded. The
nature of these shifts cannot be determined
from the limited data now available, and
it is hoped to continue the study during
the next season.

PLANETARY NEBULAE

O. C. Wilson has assembled high-disper-
sion spectra covering the lines Ha, Ni, and
N2, with the 73-inch camera (4.5 A/mm)
at the coude focus of the 100-inch, of sev-
eral of the brighter planetaries (IC 418,
NGC 6210, 6572, and 7009) in an attempt
to derive an upper limit to the kinetic tem-
peratures. He has also obtained slit spectra
of planetaries of faint surface brightness,
too faint to be observed on Mount Wilson,
with the 18-inch camera (18 A/mm) at
the coude focus of the 200-inch. This new
material suggests that the fainter nebulae
have larger expansion velocities than the
brighter ones (e.g., 7 fainter nebulae aver-
age 70.5 km/sec and 18 brighter ones aver-
age 43.7 km/sec). These velocities refer
chiefly to the Ni and N 2 lines.

Following the successful trial of a multi-
slit for the 32-inch camera at the coude
focus of the 100-inch, a similar device is
being constructed for the 6-foot camera
at the coude focus of the 200-inch. It will
be used by O. C. Wilson to photograph a
number of the brighter planetaries, and
this material, in conjunction with slitless
spectra, should provide data on the spatial
distribution of the nebular matter. Dr.
L. H. Aller will collaborate on this work.

EMISSION NEBULAE IN THE
GALACTIC SYSTEM

From an examination of 48-inch schmidt
plates centered on the galactic plane and
completely covering the range from longi-
tude 315 ° to 105 °, Sharpless has catalogued
143 galactic emission nebulae and has iden-
tified several hundred early-type stars asso-
ciated with them. Direct photographs of
a number of these nebulae were made with
the 60-inch in a search for small associated
clusterings of early-type stars that might
be used for estimates of spectroscopic paral-
laxes.

As part of an investigation of the dis-
tribution of interstellar hydrogen in the
neighborhood of the sun, Sharpless ob-
tained low-dispersion spectra of a number
of stars associated with the emission nebu-
lae IC 410, IC 1805, IC 1848, and four
others found on 48-inch plates. The spec-
tra furnished apparent distance moduli,
which, when combined with photoelectric
magnitudes and colors, will furnish paral-
laxes of the nebulae.

GLOBULAR CLUSTERS
Messier 92

The first color-magnitude diagram for
a globular cluster to be completed and pub-
lished in the new program was that for
M 92, representing a joint investigation by
Arp, Baum, and Sandage. The data
assembled are accurate magnitudes and
colors derived from photographic inter-
polations between photoelectric standards.
They range from the top of the red-giant
branch at M pv = — 3 down to a segment
of the main sequence at +4, and they
show how the subgiant branch joins the
main sequence in a globular cluster. The
variable-star gap in the horizontal branch
is conspicuous, a range in color (P — V =
+ 0.12 to +0.23) containing more than a
dozen cluster-type variables but no non-

MOUNT WILSON AND PALOMAR OBSERVATORIES

variables. Of the 7 globular clusters for
which considerable data are now available,
M 92 is the most abnormal in the sense
that the stars differ most markedly from
those in the solar neighborhood. This ab-
normality, although significant for the
study of type II populations, makes the
fitting of the diagram for the cluster to that
for stars in the solar neighborhood so am-
biguous that estimates of absolute magni-
tudes in the cluster are subject to consider-
able uncertainties.

Messier 3

In M 3, however, the stars appear to be
more normal, and Sandage, who has com-
pleted his study of the cluster, has derived
a mean absolute magnitude of cluster-type
variables, M pv = +0.1 ±0.2. This result was
obtained by fitting the subgiant branch and
its junction with the main sequence in the
diagram for the cluster and for the near-by
stars. It is considered to be reliable within
the assigned probable error, and, of course,
it establishes the absolute magnitude of the
brightest stars in M 3, in other words, of a
type II population such as that observed
in the Andromeda nebula, and hence fur-
nishes a new, reliable distance for the
spiral.

Sandage has also derived the luminosity
function for M 3 from counts on 200-inch
plates of stars within 8' from the center
and brighter than m pv : =22.5 (about 44,500
in all). He finds that 90 per cent of the
total light comes from stars brighter than
M P v= +3-8, or m pv =i9.4, but that, if the
function can be extrapolated beyond the
observable range according to van Rhijn's
function for stars near the sun, 90 per cent
of the mass of the cluster is contributed by
stars fainter than M pv = +3.0, and most of
the mass by stars too faint to be recorded.
The total mass derived in this manner is
1.4 X io 5 suns.

Analysis of the luminosity function and
the color-magnitude diagram supports the
evolutional picture for type II populations
formulated by Sandage and Schwarzschild
from computations of stellar-interior mod-
els carried out at Princeton in 1952. The
application of the theory to M 3 leads to
an age for the cluster (i.e., time since all
stars were on the main sequence) of
(5±i) Xio 9 years.

The study of luminosity functions is be-
ing extended to M 13 and M 92, and the
necessary plates with the 200-inch have
already been assembled.

Period-Luminosity Relation for Type
II Cepheids

Arp has completed his redetermination
of luminosities of type II cepheids relative
to cluster-type variables by constructing ac-
curate light-curves for 13 cepheids with
periods longer than 1 day found in the 6
globular clusters M 2, 3, 5, 10, 13, and 15.
The luminosity measures, as usual, repre-
sent photographic interpolations between
photoelectric standards furnished by Baum,
H. L. Johnson, or Whitford, some of
which are unpublished.

The curves for the 13 stars, together with
those for 6 similar cepheids in gj Centauri,
previously determined with great accuracy
by W. Chr. Martin, furnish a period-lumi-
nosity diagram in which the luminosities
in the several clusters are all adjusted to
the cluster-type variables (assumed to be
equally luminous in all the clusters). In
this diagram the cepheids appear to define
three narrow lines (residuals less than
± 0.1 mag.) separated from one another by
a factor of 2 in the periods. The line on
which a particular cepheid falls is deter-
mined by the form of its light-curve.

Interpretation of the diagram is specu-
lative at present. The light-curve might be
pictured as consisting of two or more

CARNEGIE INSTITUTION OF WASHINGTON

periods, each being one-half the next longer
period, and each representing a permitted
vibrational mode. The small differences
in the light-curve which determine the
line in the period-luminosity diagram on
which the stars fall would then indicate
which vibrational modes were dominant.

Existing spectra of these variables (ob-
tained by Joy) suggest that they correspond
to classic cepheids of one-half, one-fourth,
or one-eighth the period. For a given ab-
solute magnitude, the light-curves for the
type II cepheids which fall on the line for
the shortest periods show the closest re-
semblance to the light-curves of classic
cepheids. This picture, with its obvious
implications, may restore the continuity of
physical parameters throughout the ceph-
eid family which was recently disrupted by
a revision of —1.5 mag. in the zero point
for classic cepheids.

Long-Period Variables

Arp has also found in the globular clus-
ters a group of variables with periods of
around 100 days. They appear to be dis-
tinct from the type II cepheids in that they
fall in quite a different region of the color-
magnitude diagram but are closely allied
to Me variables in the general field. This
tentative identification opens a new field of
investigation which is being followed by
regular observations both in the clusters
studied for cepheids and in others as well.

General Properties of the Bright

Regions of Color-Magnitude

Diagrams

Color-magnitude diagrams were con-
structed by Arp for 5 of the clusters he
studied for cepheids, namely, M 2, 5, 10,
13, and 15, each diagram containing about
300 stars from the very brightest down to
absolute magnitude about +1. These di-
agrams, together with the corresponding

portions of the diagrams for M 3 and
M 92, furnish comparable data for 7 clus-
ters. After allowance for relative distances
and reddening (estimated from the posi-
tions of the cluster-type variables, the gap
in which they occur, or both), a compari-
son of the diagrams indicates several fea-
tures, some of which may have evolutional
significance. M 3 and M 92 are known to
exhibit a systematic difference in the color
indices of their giant sequences, and, in this
respect, the 5 new clusters all fall within
this observed range: M 2 and M 15 fall
near M 92; M 5 and M 13 fall near M 3;
and M 10 is intermediate. A bifurcation in
the giant sequence, near the horizontal
branch, is suggested in all the diagrams;
and in the diagram for M 13, where the
measures are most accurate, the split can
be traced nearly to the end of the giant
sequences. The horizontal branches show
great disparity in position, shape, and rela-
tive population, but there appears to be a
definite correlation between the number of
cluster-type variables and the number of
stars on the red side of the gap. A super-
position of the seven diagrams shows only
a few scattered stars brighter than the
horizontal branch and bluer than the giant
sequence, and it is in this region that the
type II cepheids are found, occupying a
roughly vertical strip above the cluster-
type variables, with color indices (at mean
light) ranging from +0.1 to +0.5 mag.
Irregular variables (some with small ampli-
tudes) fall in the giant sequence, and the
100-day variables fall along a continuation
of the bright end of the sequence.

Search for Incipient Variability among
Stars near the Variable-Star Gap

The characteristics of stars at either edge
of the variable-star gap in the color-magni-
tude diagrams for globular clusters are be-
ing investigated along two lines. Sandage

MOUNT WILSON AND PALOMAR OBSERVATORIES

has assembled a long series of blue and
yellow plates of M 3 which are now being
measured by Mr. Morton Roberts, of Occi-
dental College. This material will permit
a general study both of variable stars in the
gap and of other stars in the neighborhood.
Meanwhile, Walker has followed photo-
electrically a few individual stars (10 in
M 3 and 5 in M 5) lying at either edge of
the gap. No variability has been definitely
established (although changes of the order
of 0.05 mag. were suspected in two of the
stars in M 3), and the results suggest that
the onset of variability is quite abrupt at
the edges of the gap.

Walker also followed a number of the
brightest red giants in M 3 and M 92 but
found no evidence of variability. Both
these photoelectric programs will be carried
on through another season.

Photoelectric Measures in Messier 13

Baum has measured photoelectrically
about 30 stars in M 13, partly to establish
magnitude standards for general use, and
partly to permit a comparison, on the
basis of very accurate data, of the sequences
in the color-magnitude diagram with those
in the diagrams for M 3 and M 92 (in
which marked differences, both photoelec-
trical and spectroscopical, were reported
last year).

As an extension of this program, tenta-
tive measures of two very faint stars in
M 13 (order of magnitude 22) were ob-
tained with the new photon-counting pho-
tometer. Both stars have absolute magni-
tudes around + 7, yet their color indices
are of the order of +0.5 mag. The stars
are far too blue and faint for the main
sequence, far too red for pre- and post-
novae, and too bright for ordinary degen-
erate white dwarfs. Moreover, if they are
similar to the subdwarfs among the high-
velocity stars, their absence from a region

closer to the main sequence is puzzling.
The data are unexpected and may have
some bearing on the end product of evo-
lution in type II population.

Mass of Messier 92

Twenty-three spectrograms of 15 red
giants in M 92 have been assembled by
O. C. Wilson for the purpose, among
others, of deriving the mass of the cluster
from the dispersion in velocities of its stars.
He has now abandoned the program be-
cause of uncertainties in the measurement
and interpretation of the spectrograms aris-
ing from the abnormal weakness of metal-
lic lines. Another cluster may be substi-
tuted, because sample spectrograms suggest
that stars in M 3, M 13, and M 15 are
more nearly normal than those in M 92.
Meanwhile, the data for M 92, as they
stand, with their probable errors of about
2 km/sec for a single plate, indicate a mass
of the order of 3.3 X 10 5 suns, and, in solar
units, a coefficient mass/luminosity = 2.
This value for the mass of M 92 is of the
same order as the value, 1.4 X10 5 suns,
which Sandage derived for M 3 from its
luminosity function.

Other Studies

Among unfinished programs are two
co-operative studies by Sandage and Dr.
H. L. Johnson, of the Lowell Observatory.
One of these concerns the peculiar galactic
cluster M 67, in which Johnson has meas-
ured photoelectrically, in three colors,
about 100 stars. The data suggest a color-
magnitude diagram with strong giant and
subgiant sequence so peculiar that it may
hold clues to stellar evolution. Sandage has
obtained a series of blue and yellow plates
with the 60-inch, which he is now measur-
ing. The photographic data, calibrated by
the photoelectric, should furnish the com-

CARNEGIE INSTITUTION OF WASHINGTON

plete diagram down to the limit of the
6o-inch measures.

The second co-operative program aims
at color-magnitude diagrams in the ultra-
violet for M 3 and M 13. Sandage has ob-
tained the ultraviolet plates with the 100-
inch, using a band pass from A3300 to
A3700, and Johnson is furnishing the neces-
sary photoelectric standards on his ultra-
violet photometric system.

THEORETICAL INVESTIGATIONS

The stochastic process defined by Am-
bartsumian's model of the Milky Way
has been the object of further study by
Munch. A paper has been prepared in
which the joint probability distribution of
the brightness of the Milky Way, arising
from parts of the system extending to dif-
ferent distances from the observer, is eval-
uated.

EXTRAGALACTIC NEBULAE
Distance Scale

The unit of distance hitherto employed
in expressing extragalactic distances has
been found to be erroneous, and its value
must be roughly doubled. The order of
the revision is known to be correct, and
only minor improvements can be expected
in the future. The revision does not affect
relative distances; in other words, it does
not afreet the appearance of existing maps
of the observable region, but only the scale
on which they are drawn. These positive
statements are the results of a long pro-
gram of investigation on which progress
reports have been made over the past
several years; the fundamental significance
of the revision justifies some further re-
view of the subject.

The old unit of extragalactic distance
was based on the zero point of the period-
luminosity curve for classic cepheids. This

zero point had been derived in 191 8 (be-
fore the extragalactic regions were opened
to exploration) from the parallactic mo-
tions of a few classic cepheids in our own
stellar system. Although the value has
been under more or less constant scrutiny,
no significant revision has been found until
quite recently. Nevertheless, discrepancies
have been known for many years between
certain extragalactic data based on the zero
point of classic cepheids and corresponding
data within our own galactic system based
on another unit of distance associated with
cluster-type variables in globular clusters.
It has seemed probable that one or both of
the units were in error, and the current
programs were initiated in order to clarify
the situation. The results now confirm the
galactic unit (as to its order of magnitude)
and radically revise the extragalactic unit.
Baade, who has been largely responsible
for the investigations, describes the back-
ground and the nature of the revision as
follows.

With the recognition of the two types
of stellar population in extragalactic nebu-
lae, each containing cepheids, the first seri-
ous doubts arose as to whether the cepheids
associated with the different populations
really obeyed the same period-luminosity
relation. Two facts in particular strength-
ened these doubts: the occurrence of
cepheids with periods of less than a day
(the cluster-type variables) in population
II only, and the marked differences in the
light-curves of the cepheids of the two
populations for periods larger than 12
days. Since cluster-type variables (of type
II) had been widely used in measuring the
dimensions of our galaxy, and type I
cepheids had been used for the distances
of the nearest extragalactic nebulae, it
became of prime importance to check
whether our yardsticks in the two popu-
lations were actually the same. As a suit-

MOUNT WILSON AND P ALOMAR OBSERVATORIES 21

able test object the Andromeda nebula to an increase of all distances derived from
was selected, because it contains the two type I cepheids by a factor of about 2.
stellar populations side by side: population The inconsistency between the previous
I in the spiral arms, population II in the zero point of the cluster-type variables and
central lens and the inter-arm regions. If that of type I cepheids has been thoroughly
our present yardsticks in the two popula- confirmed by Thackeray and Wesselink,
tions were consistent, the same value for who were able to reach with the 74-inch
the distance of the Andromeda nebula Pretoria reflector the cluster-type variables
should result whether the distance indica- in the Magellanic Clouds. Again the same
tors of population I or II were used; other- discrepancy showed up, the distance mod-
wise, divergent values would be obtained. u li derived from the cluster-type variables

Investigations carried out with the 200- being 1.4 mag. larger than those from the

inch telescope during the past three years type I cepheids. That the discrepancy is

have shown conclusively that the yard- almost entirely due to an error in the

sticks with which we measured distances previous zero point of the type I cepheids

in the two stellar populations are incon- has been confirmed by Mineur and more

sistent. The distance modulus of the An- recently by Blaauw, who made use of new

dromeda nebula, as previously derived and very accurate proper motions of 15

from the cepheids of population I, is close type I cepheids provided by H. R. Morgan.

to m — M = 22.4. On the other hand, the Blaauw's value for the correction of the

distance modulus now derived from the zero point of the type I cepheids is AM =

brightest stars of population II (the clus- —1.4, in close agreement with the value

ter-type variables of the Andromeda neb- derived here and by Mineur.

ula are beyond the reach even of the 200- The new determinations of the zero

inch telescope) turned out to be m — M= points of the cluster-type variables and type

23.9. The discordance amounts, therefore, I cepheids, which of course are not yet

to 1.5 magnitudes. Either the zero point final, remove a number of disturbing in-

of the type I cepheids or that of the clus- consistencies which had become apparent

ter-type variables is in error, or both need in recent years. It is sufficient to mention

correction. only two of them: the abnormally large

In order to settle this question, Sandage size of our galaxy, which had no counter-
checked the zero point of the cluster-type part among the extragalactic nebulae and
variables by means of the color-magnitude which now falls into line with the An-
diagram of Messier 3. Using subgiants of dromeda nebula and other giant stellar
well determined trigonometric parallaxes systems, and the abnormally low age of the
to fix the zero point of the M 3 diagram, universe as derived from the red-shift
Sandage obtained for the absolute magni- constant.

tude of the cluster-type variables M pg = In addition to Baum's work on photo-
+ 0.1 ± 0.2, in close agreement with the metric standards in Selected Areas and the
previously accepted value M pg = o.o. The studies of globular clusters which estab-
discrepancy in the distance moduli of lished the unit of distance, both of which
the Andromeda nebula must therefore be are discussed earlier in this report, the pro-
almost entirely ascribed to a zero-point gram of the distance scale includes the de-
error for the type I cepheids. The indi- termination of the distance of the An-
cated correction, AM= —1.5, corresponds dromeda nebula in terms of the new unit,

CARNEGIE INSTITUTION OF WASHINGTON

in order to calibrate the many kinds of dis-
tance indicator found in the great spiral,
and the determination of distances derived
from these indicators for a sufficient num-
ber of nebulae to furnish a usable lumi-
nosity function for the nebulae themselves.
This function can then be used for sta-
tistical distances out as far as nebulae can
be recorded. The investigations may be
grouped according to whether they refer
to nebulae within the Local Group or to
more distant objects in the general field.

Studies within the Local Group

Baade has continued the study of vari-
ables in the Andromeda nebula and has
initiated a similar study of nebulae with
pure type II populations which are some-
what nearer than the Andromeda nebula.
These latter are important in their own
right, and they also serve as intermediate
steps between galactic globular clusters and
the Andromeda nebula. He reports as
follows.

During the past year a beginning was
made with the study of variable stars in
the pure type II systems of the Local
Group of nebulae. In addition to the
Sculptor System, which is being studied at
the Radcliflfe Observatory, the system in
Fornax, the two systems in Leo, and the
newly discovered system in Draco at R.A.
17 11 19.4 111 , Dec. +57° 57-7 (1950.0) are near
enough so that the cluster-type variables
can be observed at the 200-inch telescope.
All these are dwarf systems. As examples
of type II systems of intermediate bright-
ness, NGC 185 and NGC 205 were se-
lected. In these systems the cluster-type
variables are beyond the reach of the 200-
inch telescope. They are so rich in vari-
ables of higher luminosities, however, that
their investigation should supplement the
data derived from the dwarf systems in a
most desirable manner.

Miss Swope began the investigation of
the variables in the bright spiral arm at the
south-preceding end of the Andromeda
nebula and derived during the year the
light-curves of 176 variables, among which
were 123 cepheids, 14 eclipsing systems, 6
irregular variables, and 1 nova. Dr. Sergei
Gaposchkin, who spent three months as a
guest investigator at the Observatory, esti-
mated the magnitudes of the variables in
the third Andromeda field, which lies
north of Miss Swope's field and slightly
overlaps it.

Because the results thus far obtained by
Miss Swope clearly indicated that, on ac-
count of the absorption present in the in-
ner spiral arms of the Andromeda nebula,
the cepheids show a considerable scattering
with respect to the period-luminosity rela-
tion, observations of a fourth field were
started. This field, located in an outer
spiral arm, at a distance of 96 minutes of
arc south-preceding the nucleus, should be
essentially free from obscuration, since ex-
tragalactic nebulae appear in large num-
bers. The cepheids of this field should lead
to an unbiased determination of the dis-
tance of the Andromeda nebula.

Bright Irregular Variables in the
Andromeda Nebula

A study of 5 bright irregular variables in
the Andromeda nebula and M 33 by Hub-
ble and Sandage seems to define a new
class not previously recognized. The char-
acteristics include : (1) high absolute lumi-
nosity, M pg (max) = — 8.3 ±0.2 (A.D.) on
the new distance scale; (2) blue color, or
at least negligible color indices; and (3)
F-type spectra with suggestions of P Cygni
type features. Light-curves covering the
interval 1916 to 1953 show small ampli-
tudes (Am ^ 2 mag.) and irregular be-
havior. At maxima, they are among the
very brightest stars in the two spirals, and

MOUNT WILSON AND PALOMAR OBSERVATORIES

this fact together with the small dispersion
makes them useful as rough distance indi-
cators. Similar variables are known in
M 81, NGC 2403, and several members of
the Ursa Major Cloud and the Virgo
Cluster.

Novae in the Andromeda Nebula

The long-planned systematic search for
novae in the Andromeda nebula and sev-
eral other members of the Local Group
was initiated in June 1953 by Arp. Using
the 60-inch, he is assembling photographs
on every night during the entire observ-
ing season, moon and weather permitting.
Fifty plates during June and July yielded
three novae, all caught on the rise and,
consequently, carefully followed through
their maxima. Based on a preliminary
magnitude sequence, the maxima fell be-
tween m pg =i5.5 and 16. Two of the no-
vae were "slow" and the other was "fast."
Near maxima, photovisual magnitudes
were also derived. This program, carried
on for perhaps two seasons, is expected
to furnish group characteristics of normal
novae as well as to catch suitable specimens
for spectrographic investigations.

Studies beyond the Local Group

Continued observations of the M 81
group of nebulae, the Ursa Major Cloud,
and the Virgo Cluster have produced a
wealth of photometric material, but re-
duction and analysis await the final deter-
mination of standards in two of the Se-
lected Areas. Hubble and Sandage, who
are carrying out the investigations, found
only one nova in M 81 during the observ-
ing season of about five months, in con-
trast with the six or seven found in each
of the three preceding seasons.

Blue Supergiants in Type I Populations

The brightest resolved objects in two
nebulae, NGC 2903 and 4214, were studied

by Sharpless on plates taken in three colors
with the 100-inch. Photographic magni-
tudes of the bright blue stars led to a lumi-
nosity function which agrees well with the
van Rhijn function when the new scale of
distance is applied. NGC 2903 is a late-type
spiral, apparently isolated, and NGC 4214
is an irregular nebula (resembling the
Large Magellanic Cloud) in the Ursa Ma-
jor Cloud.

Total Magnitudes and Colors of
Nebulae

With the absolute unit of distance es-
tablished, the problems of determining ex-
tragalactic distances are reduced to those
of deriving relative distances from relative
apparent luminosities, first of the indi-
vidual indicators found in nebulae, and
ultimately of the nebulae themselves. The
procedure emphasizes the importance of
measures of total apparent magnitudes,
and experience has shown that total mag-
nitudes must be associated with the specific
diameters in order to have real significance.
Photoelectric measures through a series of
successively larger apertures, or some
equivalent technique, seem to offer the
most promising approach to the problem.

During the year Pettit has prepared for
publication a list of total magnitudes and
colors for about 600 nebulae, including
most of those for which red shifts are
known and many of the brightest nebulae
in the nearer clusters. Many of the older
measures have been repeated using larger
apertures, outstanding anomalies have been
re-examined, nuclei of 127 nebulae have
been newly measured; and, finally, the
color equation of the most used cell has
been reinvestigated using 54 stars in the
North Polar Sequence with known spectral
types, and was found to have remained
constant.

The nuclear colors ranged from +0.37
to +1.5 1, with 46 per cent redder than 1

24 CARNEGIE INSTITUTION OF WASHINGTON

magnitude. In general, the nuclei are red- countered in attempting to define the

der than the nebulae (only 12 per cent are "total" magnitude of a nebula on account

bluer). of the nature of its radial spread. Recent

For 150 elliptical nebulae, there appears photoelectric measures by Baum using

to be a definite correlation between color graduated concentric apertures indicate

and total magnitude, C.I. = 0.04 (m p <r — 9) without any doubt that typical nebulae

+ 0.84; the deviations may reach +0.2 extend far beyond the diameters within

mag., and four are unaccountably large, which previous magnitudes have been

Spirals are systematically bluer than ellip- measured, and that the observed magni-

tical nebulae. Subject to a very large scat- tudes are very materially dependent on the

ter, the data for spirals might be repre- degree to which outer regions are included,

sented by C.I. = 0.08 (m pg — 9) + 0.50. Any systematic dependence of inclusive-

The 19 irregular nebulae in the list are ness on distance would result in non-

not restricted to two color levels, as has linearity in the distance scale; it could

sometimes been suggested, but suggest a easily nullify the nonlinear term in the

tendency to get bluer with faintness. present red-shift relation.

Because red magnitudes in very faint No clear-cut criterion for the "total" mag-
clusters are desirable for formulating the nitude of a nebula seems to be available,
law of red shifts, and photocells do not yet As a function of angular-diaphragm diam-
reach the required limit, a jiggle camera eter d, the integrated luminosities L of
was constructed for use at the prime focus typical Eo nebulae have been found photo-
of the 200-inch. This device builds up electrically to follow the form L — Lq-\-\
smooth square images of stars and nebulae, log (d + d ), which diverges instead of
J/2, 1, 2, or 4 mm on a side, as desired. The converging to an asymptotic total. In a
plan was to derive blue, yellow, and red typical cluster it was found that the only
magnitudes, using two different sizes of limit to this relation is the blending of the
image, of the twenty or so brightest nebu- outskirts of a nebula into the outskirts of
lae in distant clusters with measured red its neighbors, and that faint nebulae are
shifts, and to calibrate the blue or yellow as extensive as brighter ones. Estimates of
magnitudes or both with photoelectric "total" magnitude in accord with this effect
measures in the hope that the calibrations of mutual limitation would depend on
could safely be extended to the red. Three- local compactness in the cluster,
color magnitudes have been measured by When integrated magnitudes are plotted
Hubble and Sandage with the jiggle cam- against log d for each of two similar nebu-
era in two of the clusters, and two-color lae, the relative displacement of the two
magnitudes in several others, but the prob- curves yields the ratio of their distances,
lems of calibration are not yet solved, without any assumptions regarding "total"
Baum, who has been investigating the diffi- magnitudes or any preselection of aper-
culties, presents a special report. tures to be used for comparative measures.

The relation between red shift and dis- The same corrections for spectral-energy

tance has thus far been investigated using shift, recession factor, and so on, must, of

"total" apparent magnitudes of comparable course, be utilized as before. Not only is

nebulae to determine their relative dis- the ratio of distances obtained, but the in-

tances. The magnitude scale itself is now trinsic similarity of the nebulae is checked

on a firm basis by virtue of photoelectric simply by noting whether the two curves

calibration, but serious difficulties are en- are alike. Intergalactic absorption and evo-

MOUNT WILSON AND PALOMAR OBSERVATORIES

lutionary fading would show up as a dis-
agreement between Am and A log d. On
the basis of present results, it is now clear
that a photoelectric procedure of this sort
will be essential to a definitive study of
the relation between red shift and distance.

Law of Red Shifts

In view of the difficulties described by
Baum, it is clear that current formulations
of the law of red shifts must be regarded
as approximations only. If, however, mag-
nitudes are derived from large diameters
or apertures, which vary more or less in-
versely with the red shifts, they should be
on a fairly uniform system, independent
of distance, and the approximation in the
law of red shifts should be correspondingly
close. Approaching the problem in this
manner, and using red shifts measured
by Humason together with the most re-
liable photometric data from all sources, in-
cluding the jiggle-camera measures, Hub-
ble finds the correlation log (c • dX/X) =
o.2mio + i.i6, where mio is the photovisual
magnitude of the tenth nebula in a cluster,
corrected for the observed red shift only
and not for recession, and dX/X is the
measured shift corrected for galactic ro-
tation only. Deviations from this relation
are very small (order of 0.1 mag.) except
in the case of the very sparse Pegasus Clus-
ter, in which the residual may be accounted
for by a size-of-sample effect. Use of the
first or third nebulae in the clusters pro-
duces a larger random scatter but no size-
of-sample effect. From the displacement
between the magnitude-red shift relations
using brightest stars in type I nebulae (us-
ing m Pg ) and the tenth nebulae in clusters
(using m P v), the difference in the absolute
magnitudes of these distance indicators is
about 11 mag. Assuming the brightest
stars to average about M pg = — 8 (new
scale), the tenth nebulae should average
about M P v= —19.

Observations of Red Shifts

Spectrograms of 78 extragalactic nebulae
were obtained by Humason at the prime
focus of the 200-inch. Many of the nebulae
are members of distant clusters and show
large red shifts. Additional members were
observed in the Ursa Major No. 1 Cluster
(R.A. n b 45 m , Dec. +55 59/) and in the
Hercules Cluster (R.A. i6 h 3™ Dec. +17°
56'), so that four red shifts are now as-
sembled in the former, and seven in the
latter. Observations of NGC 2403 were
continued, so that velocities of six emis-
sion patches as well as the nucleus are now
available for determination of the mass
from spectrographic rotation. Most of the
remaining nebulae were observed for the
purpose of improving results derived from
poor-quality spectra obtained during the
pioneer stage on Mount Wilson.

The number of nebulae observed at
Mount Wilson and Palomar is now 575.
Red shifts have already been published for
146, but many of these have since been
revised, because of improvements in the
methods of reducing the measures or be-
cause of new data. All objects for which
only one poor spectrogram was previously
available have now been reobserved.

Nomenclature of Red Shifts

In the forthcoming publication of the
data, the red shifts will be listed in the
form c • dX/X, the shifts actually measured
multiplied by the velocity of light. For
small shifts, say dX/X less than 5 per cent,
c • dX/X expresses accurately enough the
velocity of the nebulae if the shifts are
simple Doppler shifts. As dX/X increases,
however, corrections (the so-called "special-
relativity corrections") must be applied if
c • dX/X is still regarded as a simple "veloc-
ity of recession." These corrections grow
rapidly, so that, for dA/A = 2o per cent (al-
ready observed), the velocity corrected for

CARNEGIE INSTITUTION OF WASHINGTON

special relativity is only 18 per cent that
of light, and for cU/A = 30 per cent the
corresponding corrected velocity would be
only 25 per cent the velocity of light.

The use of c • dX/X in the published lists
will represent a compromise. All the meas-
ures will be corrected for galactic rotation,
which, of course, is a velocity, and the data
will be used for many problems involving
simple velocities (e.g., motions of near-by
nebulae, dispersions of velocities in groups
and clusters, etc.). On the other hand,
cosmologists are interested primarily in
dX/X, and will employ it differently ac-
cording to the particular theory they are
discussing. They are most interested in
large values of dX/X for which secondary
effects may be observed, whereas students
concerned with velocities are most inter-
ested in small values where the relativistic
corrections are unimportant. Since it is
inconvenient to list both red shifts and
corrected velocities, the form c • dX/X has
been adopted as a compromise.

Faint Envelopes and Extensions
around Nebulae

Zwicky has continued his study of
faintly luminous envelopes around nebulae,
and of extensions including bands connect-
ing widely separated nebulae. A number
of long connecting bands, between nebulae
separated by more than 4 "classic diam-
eters," have been examined, and in no
case has emission been detected. Conse-
quently, the bands are composed of stars.
Nevertheless, some of them are red and
some are blue. This same range in color
is found in the faint extensions around
M 51, where, even in so near a system, the
extensions are almost completely unre-
solved on blue and yellow plates with the
200-inch. These results suggest the ex-
istence of a third type of stellar population

whose over-all color is blue, although there
are no blue supergiants.

In one case of connecting bands between
nebulae, spectrograms have furnished sur-
prising results. This case is the triple sys-
tem IC 3481 (Sop), Anon (Sop), and IC
3483 (Sp). On direct photographs the
three nebulae are clearly connected by
long-drawn-out luminous bands. Never-
theless, spectrograms obtained by Huma-
son showed apparent velocities of the order
of +7000 km/sec for both IC 3481 and
the Anon, whereas IC 3483, fainter than
IC 3481 and with about the same diameter,
showed a velocity of only +134 km/sec.
Unless IC 3483 is a small near-by dwarf
unresolved with the 200-inch and acci-
dentally superposed on the end of a long
spur projecting out from the Anon, the
spectrographic data indicate the existence
of a very high peculiar velocity of a nebula
(by far the highest hitherto assigned, with
confidence, to any nebula) or some pecul-
iar characteristic of the law of red shifts.

INVESTIGATIONS WITH SCHMIDT
REFLECTORS

Internebular Matter

Large faintly luminous clouds have been
reported by Zwicky, spread throughout the
central regions of several rich clusters of
nebulae. One example is an elongated
asymmetrical cloud, several hundred light-
years long, within the cluster R.A. = i h o6 m ,
Dec. = —15° 40' (sometimes called the
A-Haufen). This cloud, like the exten-
sions from nebulae previously mentioned,
is red in some regions and blue in others.

A survey for variables and faint blue
stars over a field of about 40 square degrees
near the north galactic pole is being carried
out by Zwicky in an attempt to identify
stars which are clearly outside the galactic
system, in internebular space. Incomplete

MOUNT WILSON AND PALOMAR OBSERVATORIES

returns suggest that variables will average
about two per square degree between
mpg—15 and 19.5, and results of the search
for blue stars are regarded as satisfactory.

Counts of Stars and Nebulae

Zwicky has continued his counts of stars,
nebulae, and clusters of nebulae on 18-inch
and 48-inch film and plates, in the region
of the north galactic pole. On i03a-O
plates covering large areas he finds that
stars are distributed almost at random, but
that among nebulae brighter than m pg = 17,
clustering is pronounced. On both blue
and red film, average counts over large
areas in the richest fields are of the order
of 100 to 200 per square degree. The 48-
inch counts of all nebulae to the limits of
blue (i03a-O) and red (i03a-E + red fil-
ter) plates range up to 1000 and 2000 per
square degree over large areas (100 square
degrees) in the richest regions. Even in
these regions, the 48-inch counts run less
than 30 times the 18-inch counts, and
Zwicky adopts this as the expected ratio on
the assumption of uniform distribution
with no obscuration. At the galactic pole
itself, and in some regions of the Virgo
Cluster, the 48-inch counts drop to 500
nebulae per square degree, which Zwicky
considers as indicating the presence, within
the Virgo Cluster and the Coma Cluster, of
dark clouds of absorbing material. On this
interpretation, the absorption within large
clusters of nebulae such as those in Virgo,
Coma, and Corona Borealis would be of
the order of 0.5 mag.

Counts of Clusters of Nebulae

Counts of clusters of nebulae in regions
least affected by obscuration do give a
thirty-to-one ratio between the 48-inch and
the 18-inch, as theoretically expected for a
uniformly populated flat and nonexpand-

ing universe. The additional increase in
numbers of clusters as a function of de-
creasing angular size, which one should ex-
pect in the case of a real expansion of the
universe, does not seem to be present. The
fact that surprisingly few clusters have
been discovered with the 200-inch telescope
which are beyond the reach of the 48-inch
schmidt again indicates the presence of
intergalactic absorption, which beyond dis-
tances of 3 X io 8 light-years (old scale)
makes itself felt severely. Finally, Zwicky
reports that he has determined the struc-
tural indices of some very distant spherical
clusters of nebulae and finds them to be
of the same order as those for near-by
clusters.

Luminosity Function for Nebulae

A luminosity function for extragalactic
nebulae has been derived by Zwicky from
an analysis of 156 large clusters. It was
found to be exponentially declining for
nebulae fainter than M pg = — 11 (old
scale). He considers that with the dwarf
members of the Local Group, recently
found with the 48-inch, the luminosity
function for near-by nebulae is very simi-
lar to that derived from the clusters.

Progress Report of the National Geo-
graphic Society-Palomar Observa-
tory Sky Survey

During the year, 230 fields were photo-
graphed in both red and blue with the 48-
inch schmidt on the National Geographic
Society-Palomar Observatory Sky Survey.
The sky north of declination — 27 ° is now
80 per cent covered, although the coverage
is only 50 per cent with plates which meet
the very high standards set for the atlas.

Among the multitude of new objects
found on the survey plates this year are
three new members of the Local Group.

CARNEGIE INSTITUTION OF WASHINGTON

They are all dwarf stellar systems resem- now known in the Local Group. Positions

bling the Sculptor System. Their surface of the three new systems (for 1950) are :
brightnesses are very faint, but they are all R A D

well resolved on the 48-inch plates. Ap- I= -h 8m I( -s +67 18'

parent diameters range up to about i°. 151327 +0 5

Seven of these Sculptor-type systems are 17 19 24 +57 58

INSTRUMENTATION

During 1951 and 1952 photoelectric pho-
tometry at the 200-inch prime focus was
conducted by Baum using an end-on pho-
tomultiplier tube whose output current
was amplified by a sensitive DC feed-back
amplifier and recorded on a strip-chart pen
recorder. Refrigeration of the photomul-
tiplier with dry ice greatly reduced instru-
ment "noise" so that very small photo-
currents could be measured. If a star image
was bright enough to be seen distinctly in
an eyepiece, it was centered in a focal-plane
diaphragm (sometimes as small as 0.4 mm
when good seeing permitted), a color fil-
ter was interposed, and the photometer
head was swung into place behind the
diaphragm to obtain a deflection; a similar
deflection for the neighboring sky was then
obtained to determine how much back-
ground light to subtract. If a star was too
faint to be seen in the eyepiece, the pho-
tometer was placed in position by measur-
ing off its distance from a near-by brighter
star; using this procedure, stars as faint
as magnitude 21.8 were successfully pho-
tometered.

Since relatively few photons per second
are received from faint stars and since their
rate of arrival fluctuates randomly, the ac-
curacy of the DC method just described
drops off rapidly beyond magnitude 19 or
20. This situation can be materially im-
proved only by replacing the DC system
with one which integrates the total light
received (very much as a photographic
plate does) over an observation interval of
appreciable duration. For example, if one

wished to measure a 22d-magnitude star
with high precision, it would be necessary
to count the photons received from it for
an hour or more, to accumulate enough of
them for statistical accuracy.

With this end in view, the development
of a new photometer literally to count in-
dividual photons was carried out by Baum
during the past year, and the new instru-
ment was recently put into operation at
Palomar. Like its DC predecessor, the
photon-counting photometer employs a
refrigerated end-on photomultiplier tube,
focal-plane diaphragms, color filters, and
offset motions. It is also equipped with a
guide-scope so that a faint object can be
kept accurately centered in a small dia-
phragm throughout the duration of the
count. It differs from the DC system
mainly in the complexity of the electronic
equipment needed to amplify the photon
pulses, transmit them from the prime-focus
cage to the control desk, shape them, dis-
criminate them to remove "grass," and
count them. The maximum registry rate
of the system is 6000 counts per second,
corresponding approximately to a 14th-
magnitude object measured through a
typical color filter. The resolution of pulse
pairs is 5 microseconds. There is also a DC
channel which can be used for photometer-
ing stars brighter than magnitude 14 as
well as for general stand-by operation.

On June 10, sufficiently good seeing oc-
curred to attempt reaching objects at the
plate limit for the first time. An object

MOUNT WILSON AND PALOMAR OBSERVATORIES

2 9

(probably a nebula) whose presence was
indicated by the barest trace o£ an image
on a fully exposed 200-inch plate was se-
lected for test. It was photometered with-
out difficulty and found to be blue magni-
tude 23; this constitutes the first direct
measure of the photographic limit of the
200-inch telescope. The corresponding
limit for yellow plates would be about
magnitude 22.5.
Two other photoelectric photometers

were also designed and constructed during
the past year by Baum with the assistance
of Walker. Both instruments employ unre-
frigerated photomultipliers and DC ampli-
fiers. Their use will be shared by various
staff members interested in photoelectric
problems. The limit for precise work with
this equipment at the 100-inch telescope
will be about magnitude 17, beyond which
the new 200-inch photometer will gener-
ally be called upon.

GUEST INVESTIGATORS

Following the policy of previous years,
the Observatories have invited a substantial
number of guest investigators from other
institutions to make use of such telescope
time as was not required by the Observa-
tories' own staff. The following programs
have been carried out by these guest in-
vestigators.

During the fall of 1952 Dr. Lawrence
H. Aller, of the University of Michigan,
made observations for his studies of plane-
tary nebulae. In collaboration with O. C.
Wilson he obtained a number of slitless
spectrograms of these objects for the de-
termination of the isophotic contours of
the monochromatic images. The spatial
distribution of the radiating gases in IC 418
and IC 3568 was also estimated with the
aid of direct photographs taken with the
Hale telescope and loaned by Minkowski.
Studies of the spectra of the planetary
nuclei were made in collaboration with
O. C. Wilson.

Dr. Aller obtained well calibrated high-
dispersion spectra with the 100-inch coude
spectrograph of a number of B-type stars:
114 Tauri, Orionis, 22 and 42 Ononis, 15
and £ Canis Majoris, HD 36959 and HD
36960, y Pegasi, and the O-type star 10
Lacertae. These plates were taken for the
purpose of obtaining improved abun-
dances, utilizing the method of model at-

mospheres. Additional investigations were
made of the chemical abundances in the
hydrogen-poor star HD 1 60041, using
coude plates loaned by Munch.

With Minkowski, Dr. Aller discussed
the excitation of the optical emission spec-
tra of the Cassiopeia A nebulosity, which
is one of the best-known radio-frequency

sources.

The globular clusters Messier 53 and
NGC 5466 were observed with the 100-inch
reflector by Dr. James CufTey, of Indiana
University. Photometric exposures in yel-
low and blue light were obtained in order
to study as accurately as possible the rela-
tions between color index and apparent
magnitude. The material will be supple-
mented by further photoelectric and pho-
tographic observations with the 36-inch re-
flector at the Goethe Link Observatory of
Indiana University. The faint limiting
magnitude and the high resolving power
of the 100-inch were essential, however, for
a complete study of these faint, distant
objects.

Messier 53 is a typical globular cluster,
rich in stars, with a concentrated, unre-
solved nucleus; NGC 5466 is a loose clus-
ter with an easily resolved nucleus. For an
understanding of cluster phenomena, a
study of objects showing extremes in physi-
cal structure is essential, in order that the

30 CARNEGIE INSTITUTION OF WASHINGTON

features which depend on structure may observed with the 6o4nch reflector at

be distinguished from the features inde- Mount Wilson in Cio, Cu, and SA 89.

pendent of structure. Dr. T. S. Jacobsen, of the University of

Dr. Leland E. Cunningham, of the Uni- Washington, has obtained a series of spec-

versity of California, continued his series trograms of the cepheid variables v\ Aqui-

of observations on comets, satellites, and lae and h Cephei. These have been taken

unusual asteroids in September, October, with heavy exposures to bring out the de-

and November 1952, and in January, April, tails of emission features and of the sharp

and May 1953. The program of observa- interstellar cores associated with the H and

tions included the following objects dis- K lines.

covered at these Observatories : 1952c Har- Dr. Philip C. Keenan, of the Perkins

rington, 1951k P/Wolf-Harrington, 19511 Observatory, returned to the Observatories

Wilson-Harrington, 1950b Minkowksi, to continue his observations of S-type stars

1953EA Wilson, two Wilson objects found begun during the preceding year. R Gemi-

in 1951, and Jupiter X, XI, and XII. At- norum was observed during its rise to max-

tempts were made to observe during each imum in March 1953 in order to extend

month all current comets within reach, the measurements of wave lengths in both

for appearance, magnitude, and position, directions from the limits of the plates ob-

There were fewer comets than in previous tained during the preceding year. With

years, and only 7 were definitely observed, the 100-inch coude spectrograph it proved

Various other objects, including 5 unusual possible to obtain spectrograms of R Gemi-

minor planets discovered elsewhere, were norum down to M v = 9.o, with a scale of

also successfully followed. 20 A/mm, by using the spectrographic

Dr. Sergei Gaposchkin, of Harvard Col- emulsions with maximum speed in each

lege Observatory, made a detailed study region. Exposures were made on i03a-G,

of the variables in one area of the An- i03a-D, io3a-F(3), and 103-U emulsions,

dromeda nebula. For this purpose he made covering the region from 5000 to 7400 A.

estimates of the magnitudes of a substan- A preliminary inspection of the plates

tial number of variables on each of a long shows that there is relatively little change

series of plates of the area taken by Baade in the lines and bands as R Geminorum

with the Hale telescope. brightens from 8.9 m to 7.4™.

Dr. John B. Irwin and Mr. R. R. Brown- Two additional S-type stars, HD 35155

lee, of Goethe Link Observatory, made and HD 62164 (SU Monocerotis) , were

2880 two-color photoelectric observations, photographed in the green region with the

mostly with the 20-inch reflector at Palo- same dispersion in order that their radial

mar, in a program of photometric stand- velocities might be measured for the first

ards at +15°. The 8 brightest stars in the time.

Harvard C-regions C7, 8, 9, 10, 11, 12, 1, With the 18-inch camera of the Casse-

2, and 3 were observed. Numerous com- grain spectrograph of the 60-inch, the star

parisons were made with the brighter stars HD 44544 (FU Monocerotis) was photo-

of the North Polar Sequence, with the graphed in the green region, because previ-

Yerkes-McDonald photoelectric standards, ous small-scale plates had suggested that

and with the brighter stars in Selected the A5551 band of ZrO was present. The

Areas 57, 61, 68, 71, 89, 94, and 107. In new spectrogram showed, however, that

addition, sequences to 13.5 mag. were the features in this region are strong lines

MOUNT WILSON AND PALOMAR OBSERVATORIES o X

and not ZrO bands. This observation casts The program sponsored by the McMath-
further doubt on the membership o£ this Hulbert Observatory for the study of the
peculiar star in type S. infrared solar spectrum, using light col-
Additional motion pictures in color of lected by the Snow telescope on Mount
the planets Venus, Jupiter, and Saturn Wilson, continued for its fourth year. Mr.
were obtained by Dr. R. B. Leighton, of William Livingston under the direct su-
the Physics Department of the California pervision of Dr. O. Mohler made observa-
Institute of Technology. Because of poor tions on 160 days between July i, 1952 and
seeing on the nights when exposures were June 30, 1953. The limb-darkening and
made this year, results do not represent an energy-distribution measures of Dr. A. K.
appreciable advance over the photographs Pierce were continued through July 1952.
obtained in previous years. An automatic In August 1952 Mr. Livingston made
timing device has been used successfully numerous observations of the change of
to provide equally spaced exposures of the solar molecular CO lines with distance
equal exposure time. An automatic guid- from the sun's center. Investigation of
ing device is under construction, the use spot spectra in the infrared also was started
of which, it is hoped, will greatly improve at this time.

the steadiness of the image during each Dr. Mohler installed a lead-telluride cell
exposure. at the end of August. The infrared spec-
Photographs were made with the 18- trum was extended to 4.8 microns with the
inch schmidt camera by Dr. S. W. McCus- new equipment, and observations for
key, of the Case Institute of Technology, wave-length measurement and identifica-
to provide infrared magnitudes and colors tion of possible solar lines were begun. In
for late M-type stars in the area near the December 1952 a new grating was installed
galactic center. Photographs in the infra-' and applied to problems demanding high
red with Eastman Kodak iN emulsion resolution and low intensity of scattered
and a number 88a filter were taken for five light. A complete new set of east, center,
adjoining areas in longitude 320° to 340 ° and west limb tracings for improvement of
at latitude —7°. Exposures on i03a-O solar-line identification was finished. In
film were made on the same centers to May 1953, Dr. Pierce resumed his work
obtain blue magnitudes. The late M-type on limb darkening and intensity distribu-
stars were identified with objective-prism tion, using improved equipment,
spectrum plates on iN emulsion taken by At Lake Angelus Dr. Mohler continued
Nassau in Mexico. his measurement and identification of the
Infrared sequences have been established more than 6000 lines found in the Photo-
by intercomparison with Selected Area 61. metric Atlas of the Near Infra-red Solar
Measurement of the infrared magnitudes Spectrum published in 1950. All observa-
for the late M-type stars is now under way. tions for this atlas were made by means of
Preliminary results would indicate a very the Snow telescope, and it is expected that
high population, from 300 to 400 per square the companion volume will be published
degree. The limiting magnitude of the late in 1953. In June Mr. Livingston termi-
survey is about 12.5 or 13.0 in the infra- nated his residence on Mount Wilson. He
red. Measurements of the blue magnitudes was replaced by Mr. Dale Vrabec on June
are also under way, and it is hoped to have 15, 1953.

some knowledge of the colors in a few Spectrograms of a number of brighter

sample areas within a short time. red and blue stars in the cluster M 67 have

3 2

CARNEGIE INSTITUTION OF WASHINGTON

been classified by Dr. Daniel M. Popper,
of the University of California at Los
Angeles, and Dr. E. C. Yowell, of the Na-
tional Bureau of Standards at Los Angeles.
No population II characteristics are pres-
ent. Hence this cluster is not a loose
globular cluster as had been suggested, but
rather a galactic cluster with a rich giant
branch. Other observations have yielded
a preliminary value of 1.3 solar masses
for the mass of a normal Ko giant in the
eclipsing system TW Cancri. This result
tends to confirm the growing suspicion
that the "mass-luminosity" relation is valid
only for main-sequence stars.

Thirty-four spectrograms have been ob-
tained of the white dwarf 40 Eridani B,
along with sufficient plates of standard
stars so that the relativity displacement of
spectrum lines may be determined on a
sounder observational basis than hereto-
fore. The programs on H and K emission
lines in dwarf stars and on selected eclips-
ing binaries are continuing. The emphasis
in the latter program is on systems with
components above the main sequence.

High-dispersion spectra of several early-
type stars were obtained by Dr. Lyman
Spitzer, Jr., of Princeton University Ob-
servatory, in an effort either to detect the
interstellar Be 11 lines at A3100 or to show
that the amount of beryllium in the inter-
stellar gas, relative to other elements, is
less than on the earth. Analysis of the
material is not yet complete. In addition,
some 40 plates of early O and Of stars
were taken with a dispersion of 10 A/mm.
Since these stars show normal absorption
lines of measurable equivalent width, their
atmospheres can be analyzed by conven-
tional techniques; the information thus ob-
tained should reveal the structure of these
very hot atmospheres, and cast some light
on the physical conditions under which

emission lines occur in hot main-sequence
stars.

Dr. John Strong and Mr. William M.
Sinton, of Johns Hopkins University, made
preliminary measurements of planetary
and lunar temperatures using the 100-inch
telescope and a far-infrared spectrometer
developed for the purpose. Observations
were also made on the C0 2 bands at iou
in the atmosphere of Venus.

Dr. Otto Struve, of the University of
California at Berkeley, has taken an exten-
sive series of high-dispersion spectrograms
with the 100-inch coude for a study of the
atomic abundance in the components of
visual double stars. Visual binaries, both
components of which are main-sequence
stars having approximately equal lumi-
nosities and spectral types, were investi-
gated to determine whether there exists
any detectable difference in the spectra that
might give a clue to phenomena other
than those accounted for by variations in
ionization and excitation. In the case of
Y Virginis small differences in the compo-
nents were established, the most conspic-
uous difference being that the lines of one
component are wider than those of the
other. This probably represents a differ-
ence in rotation or possibly in turbulence.
No detectable difference in the abundance
of the elements was found. Though h
Equulei was not resolvable on the slit,
high-dispersion plates taken in the summer
of 1952 close to the passage through the
nodes showed resolution of the lines of
the two components in the integrated spec-
tra. The ionization and excitation condi-
tions are almost identical and the abun-
dances of the elements are the same in the
two stars. Plates have been obtained for
about a dozen other binaries, but detailed
studies of the spectrograms have not been
made.

MOUNT WILSON AND PALOMAR OBSERVATORIES

The problem of the (3 Canis Majoris
stars was also investigated by Dr. Struve.
Using series of high-dispersion plates, (3
Canis Majoris was studied to ascertain
whether the velocity curves obtained from
lines of different ionization potentials are
in phase, and whether the variation in line
width can be explained in terms of vibra-
tion. Differences in phase of as much as
17 minutes were found, but no conclusion
has been reached as to the reality of pulsa-
tions. Observations with high dispersion
were made at Mount Wilson of the F-type
variable h Scuti, which has a period of 4
hours. These were supplemented by pho-
toelectric and radial-velocity measurements
at Lick Observatory. It seems probable
that h Scuti represents a group of pulsating
stars that is detached in the H-R diagram
from the group designated as (3 Canis
Majoris variables, which form a compact
group between types Bi and B2, and have
no representatives among the later B types.
It therefore appears that there are several
regions in the H-R diagram in which pul-
sations occur and that there is no longer
reason to relate these groups to the classic
cepheids or cluster-type variables.

Studies of line profiles in Algol, X Tauri,
and a few other eclipsing binaries are be-
ing carried out by Dr. Struve in an effort
to interpret the broadening in terms of
rotation or other causes. Other investiga-
tions of line contours in O-type and early
B-type stars are in progress to determine
the relative importance of rotation and
turbulence in producing the very wide
lines often observed in these objects.

Sister Mary Therese, of Mundelein Col-
lege, has co-operated with Minkowski by
determining precise positions of the plane-
tary nebulae found on objective-prism
plates with the 10-inch telescope on Mount

Wilson and the 18-inch schmidt camera on
Palomar Mountain.

During the summer of 1952 Dr. A. E.
Whitford, of Washburn Observatory, Uni-
versity of Wisconsin, continued his study
of the colors of distant extragalactic neb-
ulae. Four late-type spirals with red shifts
of 22,000 to 27,000 km/sec were found to
be very blue in two-color photoelectric
measurements. The color excess found by
Stebbins and Whitford in elliptical nebulae
is definitely absent. This is interpreted as
the strongest support, up to now, for the
age-effect explanation of the excess in ellip-
ticals. The brightest elliptical system in
the Corona Borealis cluster (V= +22,000
km/sec) was compared photoelectrically
with M 32 in seven filter bands whose
centers ranged from 3400 to 8300 A. The
resulting differential energy curve does not
have a simple shape, but seems to indicate
that yellow giants in the distant systems
may be the cause of the two-color excess,
rather than the red giants assumed in the
first age-efTect explanation.

Dr. A. D. Code, also of Washburn Ob-
servatory, made three-color photoelectric
observations of blue supergiants in M 31 in
the fall of 1952. After elimination of in-
ternal reddening effects, the magnitudes of
the stars were found to be in best agree-
ment with a modulus of 23.8 for the sys-
tem, a value very close to that determined
from type II cepheids.

The redesigned photoelectric spectrum
scanner, built around a grating spectro-
graph, was used by Dr. Code and Dr.
Whitford to produce intensity records of
representative bright stars over the range
3400 to 10,000 A, with a resolution of about
10 A. Monochromatic magnitudes with a
precision comparable to that in filter pho-
tometry appear possible.

CARNEGIE INSTITUTION OF WASHINGTON

STAFF AND ORGANIZATION

Dr. Paul W. Merrill retired on Septem-
ber i, 1952, after over thirty-three years on
the stafT of the Observatory. Since 1946 he
has been in charge of the section of stellar
spectroscopy, and since 1950 he has been a
member of the Observatory Committee.

Most of Dr. Merrill's contributions have
been in the field of the spectroscopy of
peculiar stars with emission lines. These
contributions have included several dozen
articles on the spectra of long-period vari-
ables, which have come to be accepted as
the standard source of data on the physical
properties of these objects. With the assist-
ance of Miss Burwell he brought out the
Mount Wilson Catalog and Bibliography
of Stars of Classes B and A Whose Spectra
Have Bright Hydrogen Lines. In addition
to cataloguing 600 or 700 previously known
objects, this program has found some 900
new objects of this type through a system-
atic survey of the sky with an objective
prism fitted to a 10-inch red-corrected lens.
He has also made detailed studies of the
spectra of a substantial number of these
early-type stars showing emission lines.

Dr. Merrill has made significant contri-
butions to the study of interstellar lines.
He was among the first to investigate the
infrared spectra of stars and nebulae and
also was one of the first to apply inter-
ferometer methods to the measurement of
the separation of close double stars.

Dr. Albert G. Wilson resigned effective
June 30, 1953, to take a position at the
Lowell Observatory. Since July 1, 1949,
Dr. Wilson has been in immediate charge
of the National Geographic Society-Palo-
mar Observatory Sky Survey. Mr. George
O. Abell is replacing Dr. Wilson on the
Survey program.

Dr. Allan R. Sandage joined the stafT
of the Observatories on July 1, 1952. He
has started a program of nebular research.

Four months before his expected retire-
ment, Mr. Albert T. Mclntire, superintend-
ent of the instrument shop, died suddenly
on December 21, 1952. Mr. Mclntire came
to the Observatory in 1909 and was placed
in charge of the instrument shop in 1937.
Effective January 1, 1953, the instrument
shop has been put under the supervision
of Mr. Nichols.

Research Division

Solar Physics: Seth B. Nicholson, Edison
Pettit, Robert S. Richardson, Thomas A.
Cragg, Joseph O. Hickox.

Stellar Spectroscopy: Paul W. Merrill, Horace
W. Babcock, Ira S. Bowen, Armin J.
Deutsch, Jesse L. Greenstein, Rudolph L.
Minkowski, Guido Munch, Olin C. Wil-
son, Sylvia Burd, Mary F. Coffeen, Dorothy
S. Deutsch, Jean Humblet, Herbert L. Kyle,
A. Louise Lowen, Mildred Mathews.

Nebular Photography, Photometry, and Spec-
troscopy: Edwin P. Hubble, Walter Baade,
William A. Baum, Milton L. Humason,
Rudolph L. Minkowski, Edison Pettit,
Allan R. Sandage, Albert G. Wilson, Fritz
Zwicky, George O. Abell, Halton C. Arp,
Alice S. Beach, R. G. Harrington, Stew-
art Sharpless, Henrietta H. Swope, Merle
Walker, Paul Wild.

Secretary of the Observatory: Milton L.
Humason.

Editor and Librarian: Alexander Pogo.

Photographer: William C. Miller.

Instrument Design and Construction

Design: Bruce Rule, project engineer; Edgar
C. Nichols, chief designer; Harold S. Kin-
ney, draftsman.

Optical Shop: Don O. Hendrix, superin-
tendent; Floyd E. Day, Melvin W. Johnson,
opticians.

Instrument Shop: Albert T. Mclntire, Edgar
C. Nichols, superintendents; Oscar Swan-
son, instrument maker; Fred Scherff, Mur-

MOUNT WILSON AND PALOMAR OBSERVATORIES

doch McKenzie, machinists; Laurence E.
Blakee, electronics technician.

Maintenance and Operation

Mount Wilson Observatory and Offices
Office: Anne McConnell, administrative
assistant; Wilma J. Berkebile, secretary;
Leah M. Mutschler, stenographer and
telephone operator.
Operation: Ashel N. Beebe, superintendent
of construction; Hugh T. Couch, carpen-
ter; Kenneth E. DeHufT, engineer; John
E. Shirey, janitor and relief engineer;
Eugene L. Hancock, Alfred H. Olmstead,
Arnold T. Ratzlaff, night assistants;
Emerson W. Hartong, truck driver;

Theresa Ratzlafif, Evelyn J. Blake, stew-
ardesses; Arthur L. France, Everett W.
McConnell, Clyde Sanger, janitors.
Palomar Observatory and Robinson Labora-
tory

Office: Eleanor C. Ellison, secretary and
librarian; Dorothea Davis, secretary.

Operation: Byron Hill, superintendent,
Palomar Observatory; Benjamin B. Trax-
ler, electrician and chief night assistant;
Charles Kearns, Robert E. Sears, night
assistants; Harley C. Marshall, office man-
ager; Ferd Feryan, mechanic; Gladys
Feryan, stewardess; Ray White, power-
house operator; George W. Pettit, jani-
tor; Gus Weber, assistant mechanic.

BIBLIOGRAPHY

Abell, George O. Astronomers revise astronomi-
cal distances. Griffith Observer, vol. 17, pp.
62-71 (1953).

Abt, Arthur. An analysis of W Virginis. Read
at 1952 Victoria joint meeting of the Amer.
Astron. Soc. and Astron. Soc. Pacific; (ab-
stract) Astron. Jour., vol. 57, p. 158 (1952).

Arp, H. C. (Review) Astrophysics: The atmos-
pheres of the sun and stars, by L. H. Aller.
Pubs. A. S. P., vol. 65, pp. 166-178 (1953).

William A. Baum, and Allan R. San-

dage. The color-magnitude diagram of the
globular cluster M92. Astron. Jour., vol. 58,
pp. 4-10 (1953); Mt. W. and P. Obs. Repr.,
No. 91.

Babcock, Harold D. What's in the air? A. S. P.
Leaflet No. 291. 8 pp. (1953).

See Babcock, Horace W.

Babcock, Horace W., and Harold D. Babcock.

Mapping the magnetic fields of the sun.

Pubs. A. S. P., vol. 64, pp. 282-287 ( x 95 2 );

Mt. W. and P. Obs. Repr., No. 114.
and Sylvia Burd. The variable magnetic

field of a 2 Canum Venaticorum. Astrophys.

Jour., vol. 116, pp. 8-17 (1952); Mt. W. and

P. Obs. Repr., No. 70.

See Gjellestad, Guro.

Baum, William A. Globular clusters. I. Photo-
electric and spectroscopic observations in M3
and M92. Astron. Jour., vol. 57, pp. 222-
227 (1952); Mt. W. and P. Obs. Repr.,
No. 82.

and A. D. Code. A photometric observa-
tion of the occultation of a Arietis by Jupiter.

Astron. Jour., vol. 58, pp. 108-112 (1953);
Mt. W. and P. Obs. Repr., No. 97.
— See Arp, Halton C.

Beach, Alice. Planets from Palomar. Sci.

American, vol. 188, no. 2, pp. 17-21 (1953).
Bowen, Ira S. The spectrographic equipment of

the 200-inch Hale telescope. Astrophys. Jour.,

vol. 116, pp. 1-7 (1952) ; Mt. W. and P. Obs.

Repr., No. 71.

Optical problems at the Palomar Observ-
atory. Jour. Opt. Soc. Amer., vol. 42, pp.
795-800 (1952); Mt. W. and P. Obs. Repr.,
No. 88.

Some new tools of the astronomer. The

Halley Lecture for 1952, delivered at Oxford
on May 13. Observatory, vol. 72, pp. 129-137
(1952).

Mount Wilson and Palomar Observa-

tories. (Reports of observatories, 195 1—

1952.) Astron. Jour., vol. 57, pp. 184-185

(1952).
Burd, Sylvia. See Babcock, Horace W.
Burwell, Cora G. Classifying stars by their

spectra. A. S. P. Leaflet No. 289. 8 pp.

(i953).

Buscombe, William, and Paul W. Merrill. In-
tensities of atomic absorption lines in the
spectra of long-period variable stars. Astro-
phys. Jour., vol. 116, pp. 525-535 (1952);
Mt. W. and P. Obs. Repr., No. 79.

Code, A. D. See Baum, William A.

Cragg, Thomas. Convention of Western Ama-
teur Astronomers. Pubs. A. S. P., vol. 64, pp.
322-323 (1952).

CARNEGIE INSTITUTION OF WASHINGTON

Cragg, Thomas. Sunspot activity during 1952.
Pubs. A. S. P., vol. 65, pp. 78-80 (1953).

Deutsch, Armin J. The spectrum variable X
Serpentis. Pubs. A. S. P., vol. 64, pp. 315-317
(1952); Mt. W. and P. Obs. Repr., No. 114.

A new spectrum variable of short period,

HD 124224. Astrophys. Jour., vol. 116, pp.
536-540 (1952); Mt. W. and P. Obs. Repr.,
No. 80.

The spectrum variable 56 Arietis and

the period-line width relation. Read at
1952 Amherst meeting of Amer. Astron.
Soc; (abstract) Astron. Jour., vol. 58, p. 37

(i953).

(Review) Stars in the making, by

Cecilia Payne-Gaposchkin. Pubs. A. S. P.,

vol. 64, pp. 54-55 (1953).
Gjellestad, Guro, and Horace W. Babcock.

The magnetically variable star HD 153882.

Astrophys. Jour., vol. 117, pp. 12-20 (1953);

Mt. W. and P. Obs. Repr., No. 83.
Greenstein, Jesse L. The luminosity of the blue

star BD +28°42ii. Pubs. A. S. P., vol. 64,

pp. 256-259 (1952); Mt. W. and P. Obs.

Repr., No. 87.

High dispersion spectrophotometry of

G-type stars. Read at 1952 Victoria meeting
of Amer. Astron. Soc. and Astron. Soc.
Pacific; (abstract) Pubs. A. S. P., vol. 64, pp.
310-31 1 (1952).

A peculiarity of the spectrum of Pi

Pegasi. Astrophys. Jour., vol. 117, pp. 269-
271 (1953); Mt. W. and P. Obs. Repr.,
No. 89.

(Review) Radio astronomy, by Lovell

and Clegg. Jour. Amer. Rocket Soc, vol. 22,

P- 35 2 (1952).
Harrington, R. G. The 48-inch schmidt-type

telescope at Palomar Observatory. Pubs.

A. S. P., vol. 64, pp. 275-281 (1952) ; Mt. W.

and P. Obs. Repr., No. 115.
Joyner, Mary. See Seares, Frederick H.
Merrill, Paul W. The postminimum spectrum

of R Hydrae. Astrophys. Jour., vol. 116, pp.

18-20 (1952); Mt. W. and P. Obs. Repr.,

No. 68.

Spectroscopic observations of stars of

class S. Astrophys. Jour., vol. 116, pp. 21-26
(1952); Mt. W. and P. Obs. Repr., No. 69.

The preminimum spectrum of R Leonis.

Astrophys. Jour., vol. 116, pp. 337-343
(1952); Mt. W. and P. Obs. Repr., No. 72.

Measurements in the spectra of four

long-period variable stars of class Me. Astro-

phys. Jour., vol. 116, pp. 344-347 (195 2 );
Mt. W. and P. Obs. Repr., No. 77.

— The low-temperature spectrum of HD
45910. Astrophys. Jour., vol. 116, pp. 498-
500 (1952); Mt. W. and P. Obs. Repr.,
No. 78.

— Measurements in the spectra of ten shell
stars. Astrophys. Jour., vol. 116, pp. 501-
515 (1952); Mt. W. and P. Obs. Repr.,
No. 75.

— Oscillations in the shell star HD 33232.
Astrophys. Jour., vol. 116, pp. 516-522
(1952); Mt. W. and P. Obs. Repr., No. 81.

— Spectrograms of ten high-velocity Me
variable stars. Astrophys. Jour., vol. 116,
pp. 523-524 (1952); Mt. W. and P. Obs.
Repr., No. 76.

— The spectrum of 48 Librae from 1944
to 1952. Astrophys. Jour., vol. 117, pp. 7-1 1
(1953); Mt. W. and P. Obs. Repr., No. 84.

— Research at Mount Wilson and Palomar.
Science, vol. 116, no. 3021, prelim, p. 3
(1952).

— Emission lines in the spectra of long-
period variable stars. Jour. Roy. Astron. Soc.
Canada, vol. 46, pp. 181-190 (1952); also in
Symposium on emission-line objects, Contr.
Dominion Astrophys. Obs., No. 27, pp. 335-

344 (i953)-

— Uber neuere Arbeiten auf dem Mt. Wil-

son und auf dem Mt. Palomar. Phys. Blatter,
vol. 9, pp. 129-130 (1953)-

— Shell stars. Pubs. A. S. P., vol. 65, pp.
113-117 (1953).

— (Review) Sir James Jeans, by E. A.
Milne. Pubs. A. S. P., vol. 65, p. 97 (1953).

— See Buscombe, William.

Minkowski, R. The electron temperature in the

planetary nebula IC 418. Pubs. A. S. P., vol.

65, pp. 161-162 (1953).
Munch, Guido. Statistics of stellar color excesses.

I. Astrophys. Jour., vol. 116, pp. 575-586

(1952); Mt. W. and P. Obs. Repr., No. 73.
Interstellar lines in the spectrum of the

high-galactic-latitude star HD 93521. Pubs.

A. S. P., vol. 64, pp. 312-315 (1952) ; Mt. W.

and P. Obs. Repr., No. 115.

(Review) The astronomical universe, by

Wasley S. Krogdahl. Pubs. A. S. P., vol. 64,

p. 327 (1952).
Nicholson, Seth B. Daily solar observation at

Mt. Wilson. Griffith Observer, vol. 17, pp.

50-56 (1953).
and Oliver R. Wulf. The role of quiet

days in the mechanism of geomagnetic ac-

MOUNT WILSON AND PALOMAR OBSERVATORIES

tivity. Pubs. A. S. P., vol. 64, pp. 265-270
(1952); Mt. W. and P. Obs. Repr., No. 87.
— and others. Summary of Mount Wilson
magnetic observations of sunspots for March,
1952, to February, 1953. Pubs. A. S. P., vol.
64, pp. 205-206, 270-271, 317-319 (i95 2 );
vol. 65, pp. 51-53, 93-94, 162-164 (1953).
(Review) Comets and meteor streams,

by J. G. Porter. Jour. Amer. Rocket Soc,
vol. 23, p. 106 (1953).

Pettit, Edison. (Review) Across the space fron-
tier, ed. Cornelius Ryan. Pubs. A. S. P., vol.
64, pp. 324-327 (1952).

(Review) Kodaikanal Observatory 190 1-

1951, by the Director. Pubs. A. S. P., vol. 64,
p. 328 (1952).

and Robert S. Richardson. Motion pic-

tures of the occultation of a Arietis by Jupi-
ter on November 20, 1952. Pubs. A. S. P.,
vol. 65, pp. 91-92 (1953).

Richardson, Robert S. See Pettit, Edison.

Sandage, Allan R. The color-magnitude dia-
gram for the globular cluster M3. Astron.
Jour., vol. 58, pp. 61-75 (1953); Mt. W. and
P. Obs. Repr., No. 95.

(Review) The universe we live in, by

John Robinson. Pubs. A. S. P., vol. 65, pp.
55-56 (1953)-

See Arp, Halton C.

Sanford, Roscoe F. The spectrum and radial
velocities of W Virginis. Astrophys. Jour.,
vol. 116, pp. 331-336 (1952); Mt. W. and
P. Obs. Repr., No. 74.

Richard Hawley Tucker, 1 859-1952.

Pubs. A. S. P., vol. 65, pp. 16-18 (1953).

Seares, Frederick H., and Mary C. Joyner.
Photovisual magnitudes and color indices in
42 Kapteyn selected areas. Pubs. A. S. P.,
vol. 64, pp. 202-204 (1952); Mt. W. and P.
Obs. Repr., No. 85.

Sharpless, Stewart. The scale of the universe.
A. S. P. Leaflet No. 290. 7 pp. (1953).

(Review) Statistical astronomy, by R. J.

Trumpler and H. F. Weaver. Pubs. A. S. P.,
vol. 65, pp. 165-166 (1953).

Swope, Henrietta H. Meeting of the Eighth
General Assembly of the International As-
tronomical Union. Pubs. A. S. P., vol. 65,
pp. 11-15 (1953).

Walker, Merle F. The light-variability of Delta
Ceti. Pubs. A. S. P., vol. 65, pp. 49-51

(i953).

Wilson, Albert G. Astronomy and eschatology.
Engng. and Sci., vol. 16, no. 9, pp. 7-10

(i953)-
Wilson, O. C. Some remarks on the spatial and

kinematic structure of the planetary nebula

IC 418. Astrophys. Jour., vol. 117, pp. 264-

268 (1953); Mt. W. and P. Obs. Repr.,

No. 90.
Wulf, Oliver R. See Nicholson, Seth B.
Zwicky, F. Luminous intergalactic matter. Pubs.

A. S. P., vol. 64, pp. 242-246 (1952) ; Mt. W.

and P. Obs. Repr., No. 86.

Dispersion in the large-scale distribution

of galaxies. Pubs. A. S. P., vol. 64, pp. 247-
255 (1952); Mt. W. and P. Obs. Repr.,
No. 86.

Morphological features of the isothermal

conversion of chemical energy into propul-
sive energy. Jour. Amer. Rocket Soc, vol.
22, pp. 339-342 (1952).

Artificial meteors. Engng. and Sci., vol.

16, no. 4, pp. 20-22 (1953).

Luminous and dark formations of in-
tergalactic matter. Phys. Today, vol. 6, no. 4,
pp. 7-1 1 (1953).

Neue Methoden der kosmologischen

Forschung. Helvetica phys. acta, vol. 26, pp.
241-254 (1953).

GEOPHYSICAL LABORATORY

Washington, District of Columbia
GEORGE W. MOREY, Acting Director

The geologists whose advice and counsel this work they have adopted a technique
played so important a part in the founding which, though long used in petrology, has
of the Geophysical Laboratory and in the never been given adequate theoretical jus-
formulation of its initial program realized tification. A statistical analysis of the rela-
that the solution of the problems they met tion between cross-section area and relative
in the field required methods and tech- volume promises to provide the necessary
niques of a research laboratory. Few rock- theoretical support for the method and
forming processes can be observed in ac- results of the modal analysis determina-
tion, and the field geologist rarely is able tions ("The theory of thin-section analy-
to deduce an unambiguous solution from sis"). In addition to obtaining the abun-
only the end products of geological proc- dance data, they have extracted the feld-
esses. With a knowledge of the behavior spars, microcline and plagioclase, from
of the constituent minerals and rocks in many of these granites for the purpose
the simplified laboratory processes, how- of determining their exact composition
ever, the geologist can go to the field with ("Composition of the microcline of gran-
new insight in his search for understand- ites" and "Anorthite content of low-tem-
ing of the natural processes. In order that perature plagioclase"). The labor involved
the results of the laboratory may be most in obtaining enough sufficiently pure ma-
useful toward this end, they must be ex- terial for chemical analysis and the time
pressed in terms of the minerals and rocks required for such an analysis necessitate
which confront the field geologist. the use of other methods. Methods based

The present work of the Geophysical on optical properties ("Indices of refraction

Laboratory is aimed primarily at the un- of natural plagioclase") and the X-ray dif-

derstanding of the six major mineral fraction properties ("Lattice parameters of

groups which make up the most common the sodic plagioclases") have proved suffi-

rocks : the feldspars, quartz, pyroxenes, am- ciently accurate and precise to determine

phiboles, micas, and olivines. Perhaps the the composition and kind of plagioclase

most common rock on the surface of the feldspar.

earth is granite, which contains members With knowledge from previous cursory

of all but the last named of these major studies of the composition, kind, and abun-

mineral groups. The first step in unravel- dance of the minerals in granite, Bowen

ing the processes which yield granitic rock and Tuttle chose the simplest system repre-

is to determine the composition, kind, and sentative of granite for study from the

abundance of the constituent minerals in standpoint of synthesis. The results of

granites from various parts of the world, more than five years' work on the system

This work is being undertaken by Chayes quartz — soda feldspar — potash feldspar are

and Robbins, who have assembled a large now being prepared for publication. Many

collection of granites for which they are of the major conclusions from their study

finding the abundance of the various con- and the companion investigations relating

stituent minerals ("Modal analysis"). For to the general problem have been presented

4 o CARNEGIE INSTITUTION OF WASHINGTON

in previous annual reports. There still re- These investigators determined the liqui-

main several parts of this work, however, dus relations in the system, but the work

to be reported now. These include the na- on mixtures of high potash content was

ture of the "Residual solutions of a granite impeded by the volatilization of potash,

magma" and quantitative data on the pres- and it was difficult or impossible to identify

sures and temperatures at which natural the phases in the solidus region by optical

granites melt ("Beginning of melting of methods. The loss of potash still presents

some natural granites") . a technical problem, but Tuttle and Smith

With the information gained from the have now been able to unravel the compli-

synthetic and natural feldspar studies as cated relations in the solidus region by

well as those on the inversion character- X-ray techniques ("The system KAlSiCX —

istics of quartz, it was found possible to NaAlSi0 4 ").

establish tests of the two most probable The studies outlined above deal mainly

origins of granite. One process which may with the origin of the alkaline rocks from

lead to the production of granite is the alkaline magmas; yet the question remains,

metamorphism of sediments; the other is Where do the alkaline magmas come

the crystallization of a magma. Both proc- from ? The field evidence has led most

esses may take place, but until now no geologists to the opinion that a primordial

quantitative criteria were available to de- magma of basic composition exists from

termine whether a granite was of mag- which most other magmas arise through

matic or metamorphic origin. One clue to fractionation. This process has been dem-

the origin of a granite is to be found in onstrated many times in principle by sim-

the temperature of the high-low inversion pie systems; the skeptical geologist, how-

of quartz, which varies inversely as the ever, still asks for proof of its reality in

temperature of formation; the other clue is terms of the complex natural system of

to be found in the kind and composition many components. One step closer to re-

of the feldspars, which also depend on the ality has been taken by Schairer with his

temperature of their formation. Tuttle and work on the system K 2 0— MgO — AI2O3—

Keith subjected a granite to these tests Si0 2 . Here further evidence is found that

and found it to be of high-temperature the end product of fractional crystallization

origin ("Magmatic versus metamorphic is a granite magma even though the parent

origin of granite"). magma may be quite basic in composi-

Closely related to the granites are a large tion ("The system K2O — MgO — AI2O3 —

group of rocks called syenites which are Si02") . This is true even if these "simpli-

especially rich in alkalies. Of particular fied magmas" are contaminated by the as-

interest are the nepheline syenites, which similation of xenoliths of basic rocks, such

are well represented by the system neph- as peridotites or dunites, or of highly alu-

eline — kaliophilite — soda feldspar — potash minous sediments. Conversely, this work

feldspar. The first steps in the investiga- demonstrates that the first liquid formed

tion of these rocks were taken many years from a wide range of compositions is gra-

ago and culminated with the work of nitic. This has long been the supposition

Schairer and Bowen (1935, No. 881 1 ). of those geologists who believe some mag-

, ... 1^, mas arise from the remelting of sediments.

1 Geophysical Laboratory paper number. Other . . t

Laboratory papers will be similarly referred to Such studies ° f anhydrous systems not

by number throughout this report. only pertain to "simplified magmas," but

GEOPHYSICAL LABORATORY a T

also are the basis for fundamental studies models to illustrate the relations in some

of the metamorphic rocks. The addition of simple cases in binary systems. These

volatile components, mainly water, brings guides are useful, but, if future experi-

us very close to the natural chemical and mental work with systems containing vola-

physical environments of metamorphism. tile components is to meet with success, a

Progress in the important field of the effect far more ambitious program must be un-

of water in the formation of both the dertaken.

igneous and the metamorphic rocks has Some experimentation by Morey has
lagged behind that in the anhydrous sys- been aimed directly at the development of
terns. One reason is that the obtaining of a suitable theory for volatiles. The sys-
water pressures in experimental bombs of terns chosen involve some of the important
the order of magnitude of those which oxides of the earth's crust. For example,
might be found in natural rock-forming a simple system ("The system H 2 —
processes was not possible until Morey de- Na 2 — Si0 2 ") was investigated by Morey
veloped the method of building up pres- and Hesselgesser because the temperatures
sure by the use of an injection pump and pressures of the critical phenomena
(Year Book No. 40, 1940-1941). This not were within a range readily accessible with
only enables pressures to be developed the available equipment. It was found pos-
equivalent to the weight of ten or more sible to fix by analysis the composition of
miles of rock, but also enables those pres- the three coexisting phases — solid, gas, and
sures to be measured and held constant by liquid — and to demonstrate that under the
suitable devices. Apparatus based on this high pressures used (2500 bars) the gaseous
method has been developed by Tuttle water might dissolve as much as 33 per
(1948, No. 1 105), and by Morey and Hes- cent of sodium silicate. Such information
selgesser (1951, No. 1159; 1952, No. 1187). is of paramount importance to the devel-
With accelerating experimental progress in opment of an operational theory,
this field the theoretical treatment of sys- Other studies by Morey have dealt with
terns involving a volatile component be- systems containing water and numerous
comes of increasing importance. Much of inorganic salts. In one group studied, the
this theory is based on the works of the solubility curve is continuous from ordi-
brilliant American physicist }. Willard nary temperatures up to the melting points
Gibbs and his distinguished Dutch dis- of the salts, and points on the pressure-
ciple }. D. van der Waals, and is complex temperature curves of saturated solutions of
and little understood. The application of such salts have been determined ("Three-
the thermodynamic principles evolved by phase pressures in systems containing
them has been mainly to organic systems, water and a salt"). In one of these,
which differ radically from those which NaP0 3 , the entire solubility curve has
must be considered in geological proc- been worked out ("The system H2O —
esses. The simplifying assumptions made NaP0 3 "). In other salts the pressure-
in the organic systems are not justified in temperature curve intersects the critical
the water — silicate systems, chiefly because curve, and there is a temperature interval
of the much greater difference in volatility, in which a liquid phase is not possible.
It is therefore necessary to approach the Several such salts have been studied at
theoretical problem from a new point of temperatures above the critical temperature
view. Such a treatment has been started of water and at high pressures, and have
by Morey, who has constructed solid given puzzling results.

a 2 CARNEGIE INSTITUTION OF WASHINGTON

The compound sodium disilicate which studied in great detail, the pressure-tem-
appears in the system H 2 — Na 2 — Si0 2 perature relations of the various poly-
exhibits complex polymorphism. This aux- morphs are still not well known. Two
iliary problem has fascinated other mem- important advances in our knowledge are
bers of the staff. Kracek, Morey, and the study by Yoder (1950, No. 1137) of
England have established the relations the change in the temperature of the high-
among the eleven forms and subforms, low quartz inversion with pressures up to
and Kracek has determined some of the 10,000 bars, and a study by Tuttle and
thermal relations ("Polymorphism of so- England of the "High-quartz-tridymite
dium disilicate, Na 2 Si 2 Oo" and "Thermo- inversion" up to 1000 kilograms per square
chemistry of Na 2 Si 2 5 "). centimeter. In conjunction with these

There are other volatile constituents be- studies of the various polymorphs of silica,

sides water, such as carbon dioxide, the Smith obtained more accurate data on the

halogens (especially fluorine and chlorine), "Cell dimensions of tridymite and cristo-

boric oxide, and sulfur, which are fre- balite." As will be seen in subsequent

quently the agents of ore formation. The discussions as well, the services of an

volatile components are probably the de- X-ray crystallographer are indispensable

termining factor in the transport of ma- to the modern approach to mineralogical

terial through the medium of solubility of problems.

solid substances in the gaseous phase. Con- The pyroxene group of minerals was
firmatory evidence of gaseous solubility of tackled in the past by means of studies on
solid materials has recently been obtained binary joins relating most members of
by Morey in experiments where under the group : enstatite — f errosilite, diopside —
some circumstances the superheated steam hedenbergite, jadeite — acmite, jadeite — di-
at high pressure has dissolved almost its opside, and others. The heart of the py-
own weight of solid material ("Solubility roxene problem lies in a study of the
in superheated steam at high pressures"), system enstatite — ferrosilite — hedenberg-
In other experiments the superheated steam ite — diopside. Here may be found the
contained 7 per cent by weight of C0 2 , complex unmixing relations so often ob-
under a total pressure of 15,000 pounds served in the igneous rocks. This is a
per square inch, at 400°, 500°, and 600 ° C. tedious, long-term project from which the
In some cases, for instance with Sn0 2 and reward to be expected is as great as that
Fe 2 3 , the C0 2 greatly increased the solu- already obtained for the feldspar studies,
bility of solid in the gas; in other cases the Yoder's work on "Jadeite" and Schairer's
solubility was decreased, for instance with on the systems Na 2 — FeO — Si0 2 and
Si0 2 ("Solubility in superheated steam Na 2 — MgO — Si0 2 (see below) are con-
containing C0 2 "). Such experiments, tributions to this problem,
which are indispensable for the evolution The amphiboles are an even more com-
of the theory, could become of great value plex group than the pyroxenes, mainly be-
in the field of ore deposition. cause they contain water. It has been only

In addition to the feldspar group, which within recent years that success has been

plays the leading role in granite and most achieved in the synthesis of some of these

other rocks, there remain the major groups minerals. Bowen and Tuttle (1949, No.

quartz, pyroxene, amphibole, mica, and mi) first obtained anthophyllite under

olivine. Although silica is the most abun- what they believed to be metastable condi-

dant constituent of the crust and has been tions; the suggestion was offered, however,

GEOPHYSICAL LABORATORY 43

that anthophyllite might be stable in the AI2O3 — S1O2 — H 2 system (Yoder, 1952,
water-deficient region (see also Yoder, No. 1199). With this foundation of ex-
1952, No. 1 199). Support for this sugges- perimental data and technique available,
tion was found by Tuttle and England Yoder and Eugster felt it was timely to
("Stability relations of the amphiboles") . tackle the five-component system involving
Tremolite and riebeckite have also been a volatile constituent, K2O — MgO — AI2O3
synthesized, but their connection with the — Si0 2 — H 2 0, in which most of the corn-
related pyroxenes, diopside and acmite re- mon micas appear. In addition, with some
spectively, is not clear. It would be desir- disregard for our customary systematic ap-
able to place more emphasis on the amphi- proach to complicated mineral groups, ex-
boles in the near future, since they repre- ploratory runs were found to be feasible
sent the largest gap in our knowledge of on those micas involving six component
the rock-forming minerals. Such a hydro- oxides ("Phase relations in the micas"),
thermal program must be preceded by a Throughout the work on the synthetic
study of the anhydrous silicate systems micas, continual assistance was gained
involved, and that work, though under from studies with Smith on single crystals
way, is proceeding slowly. The work of of the natural micas ("The mica poly-
Schairer, Yoder, and Keene on "The sys- morphs").

terns Na 2 — FeO — Si0 2 and Na 2 — The sixth major mineral group, the

MgO — Si0 2 " is important for the amphi- olivines, has been studied carefully before,

boles as well as for the pyroxenes. This year, Yoder developed a method for

The micas have not been studied here- rapidly determining the composition of an

tofore in as great detail as the other major olivine by means of X-ray diffraction

mineral groups because the nature of many ("Olivine"), which will greatly aid the

of their occurrences prevents easy separa- field geologist, for example in tracing the

tion from the containing rocks, their chem- change of composition of olivine in a se-

ical composition is complicated, and they quence of olivine basalt flows,

have not yielded to previous attempts at Although the building blocks of the

experimental investigation. Their ubiqui- geologist are the minerals in the above

tous occurrence makes them of consider- major groups, there are other minerals of

able importance in studies of all rock types interest for their contribution to specific

— igneous, metamorphic, and sedimentary, theoretical problems. One of these is the

Members of the mica group are found un- zeolite called analcite. A new theory re-

der the entire range of geological condi- garding the nature of the interior of the

tions : in lavas, sediments, marbles, and earth requires a transitional layer at depths

granites; in veins and alteration zones; from 200 to 900 kilometers, between the

and as weathering products. An investi- subcrust and a uniform homogeneous

gation of their individual properties and mantle. It has been suggested that this

stability conditions was considered, there- transitional layer may result from the

fore, of great importance. The preliminary gradual change to high-pressure modifica-

work leading to the present investigation tions of the constituent silicates. Although

of the mica group began many years ago many simple compounds exhibit poly-

at the Laboratory and culminated with morphic modifications at high pressures,

the completion of the system K 2 — MgO none has previously been found in those

— AI2O3 — Si0 2 (Schairer, 1952, unpub- rock-forming minerals investigated under

lished) and the studies in the MgO— pressures up to 50,000 atmospheres. This

CARNEGIE INSTITUTION OF WASHINGTON

pressure is equivalent to about 200 kilo- Laboratory undertake to provide all the
meters depth. Failure to find the required data required for thermodynamic calcula-
transitions was considered to be due either tions of interest to the geologist. Kracek
to their being at pressures higher than and his colleagues have built up the back-
those attained experimentally as suggested ground of information required for the
by the theory, or, as investigators who op- heat-of-formation measurements ("Ther-
pose this theory prefer, to their nonex- mochemical properties of minerals"), and
istence. Yoder and Charles E. Weir, how- have data on some important minerals, but
ever, have found such a high-pressure much remains to be done,
polymorphic transition in analcite ("High- Just as thermodynamics contributes to
pressure behavior of analcite"). our understanding of the stability relations
The data on analcite were obtained be- of minerals, so do crystal-structure studies,
cause of their bearing on another problem The physical laws governing the interac-
of fundamental importance. It is desir- tion of atoms are most complicated even
able to calculate the stability range of for simple gases. Considerable progress has
minerals, particularly those which have de- been made in the mathematical theory of
fled synthesis, from their thermodynamic simple ionic crystals such as the alkali
properties. To date, only one such calcu- halides, but the theory for silicates, whose
lation of geologic interest has been made, structures are very complex, is in a rela-
that for the reaction 2 jadeite <=^ nepheline tively crude state. It is based on the hy-
+ albite. A closely analogous reaction, 2 pothesis that ions are rigid spheres which
analcite ^ nepheline + albite + 2 water, arrange themselves so that the electrical
has been accomplished experimentally and charge in any region tends to reach a mini-
the equilibrium curve determined (Yoder, mal value. The application of wave me-
1950, No. 1 123). If the thermodynamic chanics gives a first approximation to an
properties of analcite were known, in addi- accurate theory of structural energy, and it
tion to those already obtained for the other is theoretically possible to calculate values
participating phases, the equilibrium curve for the crystal energy solely from a knowl-
could be calculated for the second reaction, edge of the number and kinds of atoms in
The calculated curve could then be com- the structure. From this knowledge of the
pared with the experimentally determined crystal energy, the stability of minerals
curve and an evaluation made of the prob- could be deduced. Such information is es-
able error in thermodynamic calculations sential in explaining the natural crystalliza-
involving rock-forming minerals. Such cal- tion of minerals in terms of the simple
culations require knowledge of the entropy physical laws governing the interaction of
and volume changes taking place in the atoms. This very difficult mathematical
reaction, and this in turn requires knowl- program should be complemented by an
edge of heats of reaction, specific heat, experimental program to guide and check
specific volumes, thermal expansions, and the mathematical results. An essential part
compressibilities. The scope of the pro- of such an experimental program is the ac-
gram of the Laboratory at present does curate X-ray measurement of atomic posi-
not enable us to supply all this needed in- tions in minerals, for crystal energy is a
formation, which must be obtained, when function of atomic positions. Smith has
possible, from other sources which often made such accurate measurements on para-
are not concerned with geological ma- celsian and high-carnegieite ("Significance
terials. It is fitting, therefore, that this of the silicon-oxygen and aluminum-oxy-

GEOPHYSICAL LABORATORY

gen distances in paracelsian and high-
carnegieite") and has determined the struc-
tures of these minerals ("The crystal struc-
ture of paracelsian" and "The crystal
structure of high-carnegieite").

Much of the above-mentioned work has
centered on a specific mineral or mineral
group. Two other very significant pro-
grams of the Laboratory deal with several
associated minerals or rocks in general.
One of these is the occurrence of trace ele-
ments, those elements present in amounts
less than o.oi weight per cent. Here again
the experimenter (Eugster) has been in-
terested not merely in the accumulation
of fundamental data, but in the applica-
tion of these data to specific geological
problems. He studied the "Trace elements
in some metamorphic rocks" and found
that the particular geological environment
was clearly reflected in their distribution.
The key minor elements which provide
clues to the nature of geological processes
can be studied from an experimental point
of view, and the feasibility of such a study
is now being determined. It is proposed
that certain minor elements be added to ex-
perimentally known mineral systems with
the purpose of determining the quantita-
tive distribution of those minor elements
in the various mineral phases and in the
liquid in equilibrium with them.

The second program in this category per-
tains to the problem of determining the

"Age of rocks." The approach requires the
extraction of a suitable mineral from the
rock, the separation of certain radioactive
elements from the mineral, and the meas-
urement of the relative amounts of specific
isotopes of these elements. From knowl-
edge of the rate at which one isotope is
produced from another, an estimate of the
age of the mineral can be made. Six
methods requiring the measurement of ten
isotopes are known, but their applicability
and reliability are still to be determined.
The first step is the development and im-
provement of the methods; the second, a
comparative test of their reliability. Con-
siderable effort and a large number of
special skills are required in the execution
of this program. Although there is some
doubt whether such age measurements
will be put on a push-button basis, the
need for testing the reliability of the known
methods more than justifies the man-hours
expended. There is every hope that new
techniques will be found which will permit
the rapid determination of the age of a
rock, particularly for those rocks devoid of
a fossil record.

Other studies which have been under-
taken this year are prerequisite to future
studies or are dividends from major efforts.
These are added to the detailed discussions
of the year's work, which follow in the
order of their presentation above.

FELDSPARS, GRANITES, AND RELATED STUDIES

Modal analysis (Chayes, Robbins). Dur-
ing the summer of 1952, Chayes was able
to study and collect from some of the bet-
ter-known granite masses of Scotland,
Devon, Cornwall, western Normandy, and
eastern Brittany. Robbins made a brief
field study of the Pre-Cambrian granites
of Missouri. Thin sections of all material
collected during 1952-1953 have been pre-

pared by an improved procedure which
yields slides of so greatly improved quality
that approximately double the former area
is now available for measurement.

The theory of thin-section analysis
(Chayes). The geometrical relation be-
tween area and volume, which is funda-
mental in all the modern techniques for
thin-section analysis, seems to have received

4 6

CARNEGIE INSTITUTION OF WASHINGTON

little further attention — in the earth sci- indicate that this fraction should have

ences, at any rate — since the use of area been 20 to 40 per cent microcline, only

as a measure of volume was first suggested three of our concentrates weighed more

by A. Delesse in 1849. Delesse was aware than 1.5 gm, and less than a third of them

that his geometrical rationalization was weighed more than 1 gm. In every case,

inadequate, and presented it with appro- however, despite the rather fine grain of

priate reservations. His demonstration, many of the rocks and the characteris-

he says, applies only to rocks which are tically interstitial habit of microcline in

"homogeneous" in the sense that each of most of them, we were able to obtain at

the minerals contained is "uniformly par- least 0.1 gm of material of suitable purity,

titioned and equally developed in all direc- Difficulty in obtaining concentrates of

tions." sufficient purity to be worth detailed study

It may be shown, however, that Delesse' is probably the major obstacle to research
method is perfectly general and that esti- on the mineralogy of plutonic rocks, and
mates of relative volume based on measure- in all work of this sort methods which will
ments of relative area are not subject to yield good results on small samples are
bias from variations in grain size, shape, therefore at a premium. (For instance, in
orientation, or distribution. Each of these a recently published study on the plagio-
factors may influence the precision of a clase of plutonic rocks, Emmons and co-
result; none affects its accuracy. Chayes workers report that they examined more
has prepared the full argument for pub- than 300 specimens, of which only 30
lication. yielded concentrates of sufficient purity

Composition of the microcline of gran- and bulk.)
ites (Chayes, Robbins). We now have on A small portion of each microcline con-
hand more than 50 heavy-liquid concen- centrate was etched, stained, and counted,
trates of microcline from two-feldspar If the count indicated a microcline content
granites, and a half dozen micro- and cryp- of more than 95 per cent, the concentrate
toperthites from alkali-granites. The aver- was accepted. Materials assaying less than
age purity of the concentrates is about 98 95 per cent microcline were recleaned and
per cent, with only 12 running less than 97 reanalyzed. About half the specimens
per cent. The principal impurities are were accepted on the first count. A single
plagioclase and quartz. After a little prac- recleaning brought most of the rest above
tice, we found that the specific gravity the mark, but some required five or six
of the liquid (bromoform diluted with cleanings.

benzene) could easily be maintained with- Anorthite content of low-temperature

in the prescribed limits for intervals suffi- plagioclase (Chayes, Robbins). As plagio-

cient to permit the use of ordinary settling clase is one of the principal impurities in

tubes, and that if the powder was thor- the concentrates, proper interpretation of

oughly deslimed before treatment the set- the chemical analyses will require knowl-

tling tubes usually produced an excellent edge of the composition of the plagioclase

concentrate. in each sample. This information has been

For a number of reasons we chose to obtained for about half the concentrates

use only the 100-150-mesh fraction. Care- and should be available for the remainder

ful stage crushing of a sample of 40 gm by the end of the coming report year,

usually yields about 10 gm of powder in Since the anorthite content of our plagio-

this size. Although our modal analyses clase rarely exceeds 30 per cent, the method

GEOPHYSICAL LABORATORY

of identification proposed by Tuttle and
Bowen (1950, No. 1133) is applicable.

In this connection we have redetermined
the relation between the 20(i3i)-20(i3i)
spacing and composition, using the orig-
inal material plus several new specimens
prepared by Emmons and co-workers and
kindly made available to us by the U. S.
Geological Survey. As a result of experi-
ence with similar curve fitting for refrac-
tive indices (discussed below), we in-
cluded in the computations only samples

same specimens as above, we found for
this new spacing that An = 59.96 ( A2© ) —
62.31. Residual variance was again ex-
traordinarily small (r= 0.9994). The total
range of A2O in the new procedure is
slightly less than in the old, and as the
precision of the X-ray measurement is un-
changed (except, of course, in the region
in which the Bowen-Tuttle spacing is ob-
scured), the sensitivity of the new pro-
cedure must be somewhat less. But the
difference is slight, and for practical pur-

40i-

c
<

/ ■

*5 20

m /

■ J

*■!

1 1 1 1 1 1

1.2

1.3 1.4

A 29

1.8

(24I)-(T32)

Fig. 1. Variation of A2© (131) _ (1 3 1) with anor- Fig. 2. Variation of A2©

(241)-(132)

thite content in low-temperature plagioclase.

in which the weight percentages of anor-
thite and albite calculated from the CaO
and Na 2 of the analyses were such that
1 100— (An + Ab)|<5. For these speci-
mens the regression of composition on
spacing is An = 51. 45 (A2O) =54.60, and,
as may be seen in figure 1, variation unac-
counted for by the fitted line is very small
( r = 0.9997).

The 20 ( i3i)-20(i3i) spacing has always
yielded inferior results in the region 5<
An<i2. In the next section but one of
this report, Smith gives an explanation of
this difficulty. He has also suggested that
the difficulty might be avoided by using
the 20( 2 4i)-20(i32) spacing. Using the

with anor-
thite content in low-temperature plagioclase.

poses the methods are interchangeable.
Figure 2 shows the data and regression
line for the 20(24i)-20(l32) spacing.

The anorthite values used in both sets
of computations, incidentally, were calcu-
lated directly from the lime content listed
in each analysis; if the constants on the
right side of each equation are divided by
4.98, the "An" of the left side may be re-
placed by "CaO."

Indices of refraction of natural plagio-
clase (Chayes). A determinative chart for
the index-composition relation in this im-
portant mineral series appeared in 1952
(No. 1 177). Shortly thereafter many new
measurements were published by Emmons

4 8

CARNEGIE INSTITUTION OF WASHINGTON

and co-workers at the University of Wis-
consin in a Memoir of the Geological So-
ciety of America.

The 30 new paired index-composition
determinations presented by the Wisconsin
group increase by almost 50 per cent the
amount of such information accumulated
since the birth of optical crystallography.
It is a matter of considerable importance,
therefore, to compare the new determina-
tions with a chart based entirely on old
ones. Lack of symmetry in the number of
replications in the two sets of observations
complicates the calculations and limits in-
terpretation of their results. Only a bare
sketch of the argument and its major con-
clusions will be presented here.

For each of the 30 specimens described
by Emmons, an estimate of anorthite con-
tent (An c ) may be calculated from the
chemical analyses. These An c values re-
flect all errors and inadequacies in chem-
ical analysis, sample purification, and cal-
culation procedure. For each index deter-
mination recorded by Emmons an estimate
of anorthite was drawn from the deter-
minative chart, and the average of such
estimates for each crystal (Ann) was com-
pared with An c , as shown in figure 3. If
Ann and An e were entirely free of error
and were actually estimating the same
parameter, all the points in the figure
would lie on a straight line of unit slope
and zero intercept. As neither estimate is
errorless, none of these three conditions is
likely to be satisfied exactly. The problem
is to determine whether the observed de-
partures are large enough to indicate the
existence of significant differences between
the methods.

As the chemical analyses were not done
in replicate, a large residual variance,
which would be reflected by considerable
scatter of the points about the line, would
be virtually uninterpretable. Fortunately

the residual variance is trifling. Similarly,
the observed slope, 1.011, differs very little
from unity, and by the appropriate test
this difference fails of significance.

The observed intercept, however, is 2.62,
which is significantly larger than the theo-
retical value (zero). For the refractive in-
dices measured by the Wisconsin group,
our determinative chart indicates anorthite
values consistently lower than those calcu-
lated from the Wisconsin chemical analy-
ses. The difference is hardly large enough

50
c
<

°S

*tt

1 1

40 50

An S

Fig. 3. Comparison of An fi with An c ; Emmons
data.

to be of practical concern, but it is of con-
siderable theoretical interest; the case is
an almost ideal illustration of systematic
bias. What evidence we have suggests that
the bias is located in the new index meas-
urements rather than in the chemical anal-
yses or the determinative chart. Only two
of the specimens used by the Wisconsin
group are from plagioclase occurrences
which have been closely studied by other
mineralogists, but these two, Amelia albite
and Bakersville oligoclase, are world fa-
mous. Records compiled in connection
with development of the determinative
chart contain no less than seven sets of
measurements for Amelia and Rvt for

GEOPHYSICAL LABORATORY

Bakersville. Unless the eleven crystallog-
raphers responsible for these measurements
are wrong, the Wisconsin values for the
refractive indices of Amelia and Bakers-
ville are decidedly low.

Lattice parameters of the so die plagio-
clases (Smith). As a first step to measur-
ing the lattice parameters of all plagio-
clases, the parameters of the more abun-
dant sodic plagioclases are being measured.
Tuttle and Bowen (1950, No. 1133) have
plotted 20(i32)-20 ( i3i) against composi-
tion and demonstrated the presence of a
high- and low-temperature series. These
plagioclases are triclinic with six param-
eters, and in this study all six parameters
are being measured, instead of the one pa-
rameter measured by Tuttle and Bowen.
Selected natural materials, supplied by
Tuttle, have been measured. Prolonged
heating near the melting point (carried
out by Schairer) converted them into the
high-temperature form, whose parameters
have also been measured. Preliminary re-
sults fully confirm the variations found by
Tuttle and Bowen. There appears to be a
discontinuity near An 3 5 which confirms
the X-ray and optical evidence. One of the
triclinic angles (y # ) passes through 90°
near AnioAb 90 in the low-temperature
form. Measurements on material only par-
tially converted from the low into the
high form show that there is a contin-
uous change in the crystal structure. A
natural plagioclase of composition near
An 2 2 has lattice parameters consistent with
a position halfway between the high- and
low-temperature forms. Another one, of
composition An3o, again has parameters
consistent with an intermediate position.
Further measurements on similar plagio-
clases should throw light on their thermal
history.

Chayes has used the Tuttle and Bowen
method to measure the composition of
natural plagioclase. Accurate indexing of

the X-ray patterns shows that the (132)
reflection of Bowen and Tuttle should be
(131). Chayes had difficulty in measuring
the (131) reflection for compositions near
Anio. Examination of accurately indexed
X-ray patterns shows that the (131) peak
is only partially resolved at these composi-
tions from the peak (222). At An the
(222) peak is almost coincident with the

(131) peak, and at Ani 5 it has moved suffi-
ciently far away to eliminate confusion.
Examination of the accurately indexed X-
ray patterns suggested the use of the peaks

(132) and (241), which has eliminated the
difficulty.

Residual solutions of a granite magma
(Tuttle, Bowen). The prevalent view con-
cerning the nature of residual solutions of
crystallizing granite magmas is that a hia-
tus exists between magmatic liquids and
hydrothermal solutions. Granitic magmas
containing water and other volatiles will
tend to concentrate these constituents dur-
ing crystallization because they are not
taken up by phases crystallizing early from
the magma. The average granite has less
than 10 per cent of minerals which contain
water. For example, if a granitic magma
contained 4.5 per cent water at the outset,
the magma would consist of 50 per cent
water when 88 per cent of the liquid crys-
tallized. Although the solubility of feld-
spars and quartz is small, there are certain
silicates, such as potassium tetrasilicate, di-
silicate, and metasilicate, which exhibit con-
tinuous solubility with water. Such com-
positions would appear far removed from
the liquids to be expected from a crystal-
lizing granite magma; proof of their re-
ality, however, lies in the rocks themselves.
A compilation of the norms of rhyolites
shows that 5.4 per cent of those analyzed
have sodium and potassium metasilicates.
On the other hand, less than 1 per cent
of the granites which cooled slowly and
produced residual liquids show normative

CARNEGIE INSTITUTION OF WASHINGTON

metasilicates. There is good reason to be-
lieve, then, that a complete gradation of
liquids exists from a magma to a hydro-
thermal solution.

Beginning of melting of some natural
granites (Tuttle, Bowen). The beginning
of melting for synthetic mixtures having

1000

ported by Greig, Shepherd, and Merwin
(Year Book No. 30, 1930-1931). They ob-
served that a dry granite became half liq-
uid in one week at 800 ° C. A plausible
explanation of this difference is that the
dry granite melted metastably because the
crystalline phases present were unstable at

600

1000

2000 3000

Pressure Kg /cm 2

4000

Fig. 4. Curves showing melting behavior of two natural granites together with the ternary minimum

compositions closely approaching those of
the average granite has been determined.
In order to gain some knowledge of the
effect of other possible constituents not
present in the synthetic system, the melt-
ing behaviors of two natural granites were
studied. The results are shown in figure 4.
If the curve is projected to zero water
vapor pressure, anhydrous granites will be-
gin to melt at about 950°. This tempera-
ture is about 1 50° higher than that re-

the temperature of the experiment. That is,
the granite consisted essentially of quartz,
microcline, and plagioclase, whereas the
stable assemblage at 800 ° C would be
quartz and sanidine. In the present work
the primary phase of crystallization in both
the natural granites examined is biotite.

Magmatic versus metamorphic origin of
granite (Tuttle, Keith). The continuing
controversy regarding the origin of gran-
ite is largely the result of conflicting field

GEOPHYSICAL LABORATORY

evidence. The present state of the prob- In order to test these tools a young gran-

lem is testimony to the inadequacy of field ite was required which had been quickly

relations alone as a source of decisive evi- cooled and which had not suffered subse-

dence for determining whether a particu- quent recrystallization. The Beinn

++ +W ++

QUARTZ FROM RHYOLITE •
QUARTZ FROM GRANITE +

574

INVERSION BREAK ON COOLING °C

Fig. 5. Inversion temperature of quartz formed at high and low temperatures in similar chemical
environments, M, granites from three localities in Maine: North Sullivan, Jonesport, and Wallace
Cove. P, Pikes Peak granite. W , granite from Westwood, Massachusetts. S, White Silver Plume
granite, Colorado.

lar granite was produced by the meta-
morphism of sediments or by the crystal-
lization of a liquid magma. Two new
measuring sticks have recently been devel-
oped which can be brought to bear on this
problem. One makes use of the variable
inversion in quartz, and the other, the
polymorphism of the feldspars.

Dubhaich granite of the island of Skye
appears to satisfy these requirements.

It has been shown that the high-low in-
version of quartz grown in similar chemi-
cal environments varies inversely as the
temperature of formation (Keith and Tut-
tle, 1952, No. 1 183). That is, the quartz
from a rock formed at high temperatures

CARNEGIE INSTITUTION OF WASHINGTON

will exhibit a lower inversion temperature temperature, it is concluded that the Skye

than quartz formed at a low temperature, granite is the result of primary crystalliza-

if the chemical environments are similar tion from a magma.

(fig. 5). The results of the inversion tern- The results from the second measuring

perature measurements of the quartz from stick, the polymorphism of the feldspars,

INVERSION BREAK ON COOLING °C

Fig. 6. Results of inversion temperature measurements of the quartz from Skye granite

the Skye granite are given in figure 6.
They indicate that the quartz from the
Skye granite is different from the quartz
of most granites previously studied and
resembles, in inversion characteristics, the
quartz from rhyolites. It is generally
agreed that rhyolites are formed at high
temperatures from a liquid magma. There-
fore, on the basis of the quartz inversion

confirm the results from the study of
quartz. The compositions of both the pot-
ash and plagioclase feldspars were deter-
mined by means of an X-ray method de-
scribed by Tuttle and Bowen (1950, No.
1 133). In addition, the optic axial angle
(2V) for each specimen was carefully
measured on a universal-stage microscope.
Since 2V is known for a specific composi-

GEOPHYSICAL LABORATORY

tion of feldspar and for each polymorph of
that composition, the polymorph could be
determined for the Skye granite feldspars.

The potash feldspars showed a consider-
able range in iV ' , but some were close to
sanidine, the high-temperature polymorph.
The variation in iV is attributed to the
partial inversion to low-temperature forms.
The plagioclase feldspar measurements fall
on the curve for high-temperature plagio-
clases. It is therefore concluded that the
feldspars in the Skye granite formed at
high temperatures.

These studies on the minerals found in
the Skye granite clearly point to an origin
at high temperatures. In fact, this particu-
lar granite appears to be a missing link be-
tween granites and rhyolites. Of still
greater importance is the development of
methods which can distinguish between a
granite formed from the metamorphism of
a sediment and a granite crystallized from
a magma.

The system KAlSiOi—NaAlSiO* (Tut-
tle, Smith). The system has been studied
by both dry and hydrothermal methods.
The phases found are given in table i . The
stability relations (fig. 7) have been
worked out in the subsolidus region em-
bracing all compositions up to 1000 ° C.
Phase-equilibrium relations in composi-
tions enclosing more than 40 per cent of

the NaAlSi0 4 molecule have been estab-
lished only up to the liquidus.

Perhaps the most important result of our
study is the location of a miscibility gap be-
tween the two phases kalsilite and low-
nepheline, below approximately 1000 ° C.
The composition of the coexisting phases
in the miscibility gap has been measured
by X-ray methods. One of the most in-
teresting features of this unmixing was the
rapidity of the process. For example,
when a single phase quenched from a high
temperature was heated in an X-ray cam-
era at 500° C, unmixing began within five
minutes. Accordingly, measurements on
the top of the miscibility gap are uncertain,
for some unmixing may have occurred
during quenching.

It is not surprising that kalsilite is the
common modification of KAlSi0 4 found
in the volcanic rocks, for this is the stable
form in the temperature range expected in
these rocks. The inversion temperature of
kalsilite at atmospheric pressure is 840° C
in pure KAlSi04 and rises rapidly with
increasing solid solution of NaAlSi04. Kal-
iophilite, found in ejected blocks at Monte
Somma, Vesuvius, is apparently different
from all the synthetic phases encountered
in this study.

The solidus (low-nepheline + liquid),
the curve representing the composition of

TABLE 1
Phases in the KAlSi0 4 — NaAlSi0 4 system

Symbol Composition Name

H2 NaAlSi0 4 Low-nepheline

H20 NaAlSi0 4 High-nepheline

LC NaAlSi0 4 Low-carnegieite

HC NaAlSi0 4 High-carnegieite

HI KAlSi0 4 Kalsilite

01 KAlSiO, "Orthorhombic"

02 K 4 / 5 Nai/ 5 AlSi0 4 "High-orthorhombic"

H4 K 3 / 4 Nai/ 4 AlSi0 4 "Double-nepheline"

K KAlSi0 4 Kaliophilite

* The cell dimensions vary with composition and temperature.

Symmetry

Cell dimensions (A)*

Hexagonal

Orthorhombic

Triclinic?

Cubic

Hexagonal

Orthorhombic

Orthorhombic?

Hexagonal

a, 10.0; c, 8.4

a, 10.2; b, 17.6; c, 8.5

a, 7.32

a, 5.16; c, 8.71

a, 9.1; c, 8.56

a, 20.5; c, 8.5
a, 26.9; c, 8.5

CARNEGIE INSTITUTION OF WASHINGTON

crystals in equilibrium with liquid, has
been determined from 30 to 70 per cent
NaAlSi0 4 by measuring the composition
of low-nepheline from the lattice param-
eters. Above 1520 ° C, a new phase, "H4,"
in equilibrium with liquid has been found,
giving a phase boundary between (low-

lized dry at 1150° C. Below this tempera-
ture it is hexagonal, and above it the sym-
metry is lowered to orthorhombic. The
inversion is rapid and of the displacive
type. A small heat effect was observed at
850° C, on the differential thermal analy-
sis apparatus. Low-nepheline made hydro-

600-

400-

NaAISiO.

KAISi0 4

Fig. 7. Stability relations in the KAlSi0 4 — NaAlSi0 4 system. Hi, kalsilite; H2, low-nepheline;
H2O, high-nepheline; H4, double-nepheline; Oi, orthorhombic; O2, high-orthorhombic; HC, high-
carnegieite.

nepheline + liquid) and ("H4" + liq-
uid) . The stability range of this new phase
has not yet been determined. The region
of high potash content and high tempera-
ture has turned out to be very complex,
and in this region a new phase, "O2," has
been discovered near 20 per cent NaAlSi0 4
and 1500 ° C.

An inversion has been found in low-
nepheline of composition 100 per cent
NaAlSiCX at 850 ° C, for material crystal-

thermally at 800 ° C also shows this inver-
sion, but at the higher temperature of ap-
proximately iooo° C. Its lattice parameters
are also slightly different at room tempera-
ture, but the reason for this discrepancy
has not yet been established. No inversion
has so far been found in low-nepheline of
adjacent compositions.

Lattice parameters have been deter-
mined for the nepheline-kalsilite solid solu-
tions as a function of composition. The re-

GEOPHYSICAL LABORATORY

suit for the a spacing is shown in figure 8.
The results have proved useful in estimat-
ing the composition of natural nepheline-
kalsilite solid solutions. The presence of
excess silica (albite) in solid solution ap-
parently does not greatly influence the de-
termination of the ratio of KAlSi0 4 to
NaAlSiQ 4 .

HI (2a)

10-30-

10-20-

10 10 —

1000-

9-90-

V H4 (o /2 )

H2 (a)

KAISiO,

1:3

NaAISiO.

NaAISL0 o

o o

Fig. 8. Variation of the lattice parameter a
with composition for the phases Hi, H2, and H4
in the KAlSi0 4 — NaAlSi0 4 system. The change
of a with composition for the H2 phase in the
NaAlSi0 4 — NaAlSi 3 O g system is also shown for
compositions having 10, 20, and 25 per cent
NaAlSi 3 O g .

Another interesting result of the study is
the sharp change of slope in the lattice
parameters a and c of low-nepheline when
plotted against alkali ratio at room temper-
ature. This discontinuity is at a ratio cor-
responding to 3 sodium atoms and 1 potas-
sium atom. There is also a break between
90 per cent and 100 per cent NaAlSi0 4 ,
presumably related to the inversion in
pure NaAlSi0 4 . These results can be in-

terpreted in terms of the crystal structure
determined by Buerger and Hamberger.
The silicon and aluminum tetrahedra form
a framework (like that in tridymite) with
the alkali atoms filling holes in the frame-
work. Three of the holes are small and
one is large. As the sodium atom is small
and the potassium atom large, the ideal
structure contains 3 sodium and 1 potas-
sium atoms. If more potassium atoms en-
ter the lattice (with corresponding loss of
sodium atoms), they must enter some of
the small holes and thereby tend to force
the lattice apart. This may explain the
increase of lattice parameters found experi-
mentally. If extra sodium atoms enter the
lattice, small atoms enter large holes and
the lattice collapses, thus giving the re-
duced parameters found experimentally.
When large potassium atoms enter small
holes the lattice is forced to expand, but
when the small sodium atoms enter large
holes the natural rigidity of the tetrahedral
framework partly compensates for the
small atoms. This explains the observed
change of gradient of lattice parameters
versus composition. When sufficient small
atoms enter large holes in the lattice, the
lattice can collapse to a configuration of
lower symmetry, a fact which affords an
explanation of the reduction of symmetry
only near pure NaAlSi0 4 . The substitu-
tion of albite in pure NaAlSi0 4 stabilizes
the lattice again, presumably because the
extra silicon atoms change the framework
so that the sodium atoms again fill the
holes.

When most of the sodium atoms in low-
nepheline are replaced by potassium atoms,
the slightly kinked tetrahedral framework
straightens out to give the kalsilite struc-
ture. This reduces the lattice parameter a
from 10 A to 5 A, but the structure is very
similar to that of low-nepheline. The
"H4" structure is intermediate between
kalsilite and low-nepheline and probably

CARNEGIE INSTITUTION OF WASHINGTON

represents the first stage in the straighten- spinel has a ternary eutectic with leucite,

ing out of the low-nepheline structure. At corundum, and spinel as the crystalline

800 ° C, the break of the lattice parameter phases in equilibrium with liquid at 1543°

with the 3:1 atom ratio in low-nepheline ±5° C at the composition leucite 88.0,

has almost disappeared, showing that corundum 1.5, spinel 10.5 per cent by

either (a) the holes now have the same weight.

size, (b) the sodium and potassium atoms Quenching-run data were completed for

have the same size, or (c) the sodium and that portion of the join leucite — forster-

potassium atoms are disordered. It is in- ite — periclase whose liquidus temperatures

teresting that Donnay and Donnay sug- lie below 1650° C, the practical upper limit

gested that mechanism (b) was responsible for platinum-wound resistance furnaces,

for the change in the monoclinic-triclinic The primary phase volumes of periclase,

inversion with temperature in the alkali forsterite, and leucite are cut by this join,

feldspars. but no piercing points of quaternary uni-

The system K 2 — MgO — AI2O3 — Si0 2 variant lines lie on it. The composition

(Schairer). The three joins leucite — for- K 2 • 6MgO ■ AI2O3 * 6Si0 2 (the mica

sterite — spinel, leucite — corundum — spinel, phlogopite is K2O ■ 6MgO • AI2O3 * 6Si0 2 *

and leucite — forsterite — periclase were com- 2H2O) lies on this join; and this anhy-

pleted during the past year but have not drous composition has a forsterite liquidus

yet been written up for publication. The at i628°±5°, forsterite crystals are joined

first two of these joins are ternary sys- by leucite crystals at 1478^5°, and the

terns within the quaternary system K 2 — composition begins to melt at 1428° ±5° C.

MgO — AI2O3 — Si0 2 . The system leucite — A new join, leucite — forsterite — K 2 0*

forsterite — spinel has a ternary eutectic AUOs^SiC^, has just been started, and

with leucite, forsterite, and spinel as the the first three compositions are now in

crystalline phases in equilibrium with liq- course of preparation. This join is of im-

uid at 1463 ° ±5° at the composition leucite mediate interest to Yoder and Eugster,

75.5, forsterite 17.5, spinel 7.0 per cent by who are studying the relations between

weight. The system leucite — corundum — phlogopite, muscovite, and clinochlore.

HYDROTHERMAL AND AUXILIARY STUDIES

The system H 2 — Na 2 — Si0 2 (Morey
and Hesselgesser) . Work on the 500° and
600 ° isotherms in this system is complete,
except for the high-Na 2 part. A plati-
num-lined bomb should soon be available,
with which we hope to extend the studies
at 400 °, 500 °, and 600 ° C far enough to
reach the alkali-rich boundary of the field
of Na 2 Si0 3 .

Numerous experiments have been made
at the critical temperature of water, 374 ° C,
to establish the boundaries of the first crit-
ical end-point curve, and some additional
work has been done at lower temperatures.

It appears certain that both the first and
the second critical end-point curves lie be-
tween 374° and 400 ° and that the pressures
at the second critical end-point curve are
very high, greater than 3000 bars. Within
this short temperature interval of less than
26°, the vapor pressures of solutions in
equilibrium with crystalline solids increase
at least 15-fold. Moreover, other studies
(1952, No. 1 187) indicate that this may be
a common relationship in systems contain-
ing water and salts of low vapor pressure.
In addition, some experiments have been
made at lower temperatures to tie in with

GEOPHYSICAL LABORATORY ^

work of others at temperatures below equilibrium G + L + S, which, provided

ioo° C. It is hoped to assemble all this in- no compounds are formed, goes from the

formation in one review paper on the eutectic G + L + ice + S to the melting

entire system from room temperature to point of the salt, is intersected by the crit-

the melting point of some of the anhy- ical curve which joins the critical point of

drous compounds. water with that of the salt. There must be

Three-phase pressures in systems con- two such intersections, P and Q, giving rise

taining water and a salt (Morey). The to a lower and an upper critical end point,

determination of three-phase pressures gas and dividing the P-T curve into three

+ liquid + solid in binary water-salt sys- sections. The first section is from low

terns, that is, the vapor pressures of solu- temperatures up to the first or lower criti-

tions saturated with the salt in question, cal end point, P; the third section is from

has been continued, and has produced the melting point of the salt down to the

some puzzling results. The salts studied second or upper critical end point, 0; and

may be listed in two groups : the second section, between the two critical

i. Salts whose solubility curve does not end points, is unstable except just adjacent

intersect the critical curve. The solubility to the critical end points, where it is meta-

curve is continuous up to the melting point stable.

of the salt, and the three-phase pressure at The following salts have been found
the critical curve of water is less than the to have a first critical end point: LiF,
critical pressure of water. The following NaF, Li 2 S0 4 , Na 2 S0 4 , T1 2 S0 4 , Li 2 C0 3 ,
salts are included in this group : NaCl, KF, Na 2 C0 3 . Points on their three-phase
KC1, CsCl, PbCl 2 , Na 2 0'2B 2 3 , Na 2 0* curves at 400° and 500° C have been de-
3B2O3, Na 2 0'4B 2 3 , NaP0 3 , KP0 3 , termined for T1 2 S0 4 , Li 2 S0 4 , K 2 SC> 4 , and
Na 4 P 2 7 , K 4 P 2 7 , K 2 C0 3 . Three-phase a point at 400° (800 bars) for Na 2 S0 4 .
pressures in these systems have been deter- These pressures are greater than the ex-
mined at 400 °, 500°, 600 °, and, when nee- trapolated P-T curve of water, and in each
essary to outline the entire pressure-tem- case the slope of the P-T curve is positive,
perature curve, at 700° C. It is difficult to reconcile these P-T curves

2. Salts whose solubility curve intersects with the requirements of a system having

the critical curve. The test for this condi- end points of the P-Q type. In some of

tion is that at the critical point of water, these systems, complications arise owing to

374° C, and under the critical pressure formation of two immiscible liquid layers,

(221.06 bars = 3206.17 psi) or slightly a condition which gives rise to formidable

higher (241 bars = 3500 psi), water theoretical complications. Also, solubilities

(steam) does not melt the salt; hence the of salt in gaseous water of over 50 per cent

vapor pressure of the saturated salt solu- have been observed. The applicable theory

tion must be greater than that of water at is being developed,

its critical point. The system H 2 0—NaP0 3 (Morey). To

This is the condition for a first critical one interested in silicates, the chemistry

end point. Probably the simplest case in of the phosphates also is of interest. In

which there is a first critical end point both cases the structural element is a tetra-

is that in which there is also a second hedral grouping of four oxygen atoms

critical end point without complication around a central positive atom, and the

from the formation of two immiscible liq- various compounds are formed by repeated

uids. In this case, the P-T curve for the polymerization. The three-phase composi-

CARNEGIE INSTITUTION OF WASHINGTON

tions, pressures, and temperatures have
been determined from ioo° C up to the
melting point of NaP0 3 (629° C). The
solid phases are: NaH 2 P0 4 , incongruent
melting point 169°, 4.1 bars; Na 2 H 2 P 2 C>7,
incongruent melting point 375°, 12.5
bars; NaP0 3 II, transition temperature to
NaP0 3 I 443°, 15 bars. This is one of the
few complete solubility curves of water-salt
systems which have been determined.

Polymorphism of sodium disilicate,
Na 2 Si 2 5 (Kracek, Morey, England).
From results published in earlier papers by
this Laboratory, it has become evident that
sodium disilicate exhibits polymorphism;
in particular, it was found that materials
crystallized dry had different optical prop-
erties from those crystallized in bombs in
the presence of water, and that the prod-
ucts of the dry heat treatment exhibited
two readily reversible rapid transitions.
Further work showed different X-ray dif-
fraction patterns for material crystallized
under different conditions. Some of the
different forms were found to be mutually
convertible, with speeds of inversion de-
pendent on the temperature. In addition,
it was found that two of these inert forms
have rapid, readily reversible transitions.

The principal inert forms or phases cor-
respond by analogy to such slowly con-
vertible phases as the quartz, tridymite,
and cristobalite forms of silica, or the neph-
elite and carnegieite forms of NaAlSi0 4 .
As such they could be given mineral
names; but since in nature, in the presence
of water, they would be unstable with re-
spect to other mineral assemblages present
in rocks, they could not properly be re-
garded as even artificial minerals. Indeed,
the only mineral of disilicate composition
containing sodium listed in Dana's Text-
boo^ of Mineralogy is rivaite, (Ca, Na 2 )
S12O5.

We have given the different inert forms
of Na 2 Si 2 5 capital-letter names, A, B, C,

etc., A being the phase stable at the highest
temperature. The modifications produced
as a result of rapid reversible transitions in
the principal phases are analogously given
lower-case Greek-letter names combined
with the name of the principal phase with-
in which they occur, thus Aa, A3, etc., a
being the high-temperature modification.

The transformations of the principal
forms were studied both hydrothermally
and under dry conditions. The transfor-
mations in the dry state were found in
general to be more sluggish and to require
higher temperatures for conversion than
the hydrothermal ones.

The relations of the various polymorphs
are given schematically in figure 9. Phase
A was produced by crystallizing a pow-
dered glass of the requisite composition
above 670 ° C. Phase B was also produced
in the dry way within the temperature
range 591 ° to 650° C. Hydrothermally, B
was obtained at 430° C and 6 atm water
vapor pressure, as well as at lower temper-
atures and higher pressures. Phase C was
not obtained in the dry way, but was pro-
duced from a glass, A or B, under 1500
and 3000 psi water vapor pressure (see
inset in fig. 9) at temperatures near 660 ° C.
Phase D was not obtained in the present
experiments, but had been identified by
Morey and Yoder in previous work. It
was found to transform to A at 722 ° C.
Phase E was obtained only by heating
glass in the temperature range 500 ° to
588 ° C. Phase F was discovered by H. S.
Roberts in the course of work on the latent
heats of various silicates. He observed that
Na 2 Si 2 5 when crystallized below about
700° always gave the two rapid transitions
characteristic of phase A. If, however, the
preparation was first heated above the
melting point of A, and then crystallized
above 706 °, it usually appeared in a new
variety which exhibited neither of the two
rapid transitions, and melted several de-

GEOPHYSICAL LABORATORY

grees lower than the stable phase, with
a higher latent heat of melting. When
partly melted, it often refroze and then
gave a second melting point, that of A.
Later work has confirmed Roberts' state-

Thermochemistry of Na 2 Si 2 5 (Kra-
cek). Thermochemical data for sodium
disilicate are included in table 6 with those
for other alkali silicates to be described in
a following section. Table 2 presents a

600

700

1000 2000 3000 psi

Fig. 9. Imaginary vapor-pressure curves for the various phases of Na 2 Si 2 5 . Solid lines represent
stable states and dotted lines metastable states at the various temperatures t° C. The position of the
curves indicates the relative stability of the phases or their modifications, although their actual posi-
tions or shapes are unknown. Those for phases A, B, and E were chosen from thermochemical
considerations; those for C, D, and F were chosen arbitrarily. The inset figure represents the results
of experiments on the formation of phase C at the indicated pressures of water vapor.

ments in general, and shows that F when
cooled to room temperature yields an X-ray
pattern similar to that of A a taken at 820°,
but noticeably simpler. F cannot be re-
tained long at room temperature; it inverts
in about a day to Ay. Present data indicate
that F is metastable with respect to A and
monotropically related to Aa.

more detailed analysis of the values of the
heats of transformation and the informa-
tion that can be derived from them con-
cerning the stabilities of the various phases.
It will be noted that form E has the lowest
heat of crystallization, measured at 25 ° C,
of the three forms studied; also that if the
reactions would take place at 25 , E as

CARNEGIE INSTITUTION OF WASHINGTON

well as A would transform to B with evo-
lution of heat, and hence that both E and
A are metastable with respect to B at 25°.
It is further seen that E is also metastable
with respect to A at 25 °, and, indeed, it
was found in the dry-heat-treatment studies
of E that in some cases E was at least par-
tially transformed to A, with the remain-
der simultaneously transformed to B at
temperatures at which A is metastable
with respect to B. The preferential crystal-

TABLE 2

Thermochemistry of polymorphous forms of
sodium disilicate

Latent heats of phase change
Condi- AH (sol, 74.7° C) at 25° C (kcal/mole)

tion (kcal/mole)

gl > x E > x A > B

Glass.. -104.70

-2.72

E -101.98

-4.66 -1.94

A -100.04

-5.25 -2.53 -0.59
B - 99.45

gl, glass; x, crystal; A, B, E, polymorphous
forms. No measurements were made on phases
C, D, and F.

lization of glass to E at certain tempera-
tures rather than to the stable B is a clear
instance of the operation of the Ostwald
stepwise rule for phase transitions.

Solubility in superheated steam at high
pressures (Morey). This work has been
continued, and some of the results are as
follows :

CaCOs: Solubility at 500 C, 15,000
psi = 12 X io -6 weight fraction. At the
same time excess C0 2 was carried over in
amount 9X io~ 6 .

BeO: 400°, 15,000 psi = 24X10" 6 ; 500°,
15,000 psi = 55 X io -6 .

In addition, work has been done with
germanium oxide, which shows a much
larger solubility than quartz, and with
uranium oxide.

Solubility in superheated steam contain-
ing C0 2 (Morey). Carbon dioxide was in-
troduced into the system by saturating the
water with CO2 under the pressure of
liquid C0 2 at room temperature. This
gives about 7 per cent C0 2 . Much trouble
resulted because of the corrosive action of
the carbonic acid on iron. The water satu-
rated with C0 2 corroded the valves of the
pump in a short time, and the water pass-
ing into the bomb carried a large amount
of ferrous carbonate. The present method
is to fill a stainless steel cylinder one half
with water and the other half with liquid
CO2. This cylinder is between the pump
and the bomb, so that the valves do not
come into contact with the C0 2 -saturated
water. The water, saturated with CO2 and
under pressure, passes through a small
bomb containing an ion-exchange resin to
remove the last trace of iron just before it
goes into the main bomb. These steps have
so greatly improved the operation that it
is desirable to repeat all previous runs.

Following are the preliminary results for
solubility at 500° C, 15,000 psi total pres-
sure, with about 7 per cent C0 2 present :

CaC0 2 : 940 X 1 o~ 6 , a more than 7-fold
increase as the result of the presence of

co 2 .

Cassiterite: 50X10" 6 , an increase of 20-
fold.

Hematite: 230 Xio -6 , an increase of 3-
fold.

Quartz: 19X10" 4 , a decrease in solu-
bility of 6 X io -4 . •

Albite: The results are not satisfactory;
but it appears probable that the CO2 de-
creases both the silica and the alumina.

It is planned to keep this program in
active operation. More work should be
done with inorganic oxides; the necessary
changes should be made for working with
more soluble salts such as K 2 SC>4, and
also for working with molten salts, such
as NaCl, at pressures greater than the
three-phase pressure.

GEOPHYSICAL LABORATORY

QUARTZ, PYROXENE, AND AMPHIBOLE GROUPS

High-quartz-tridymite inversion (Tut-
tle, England). Data have been obtained
on the high-quartz-tridymite inversion up
to iooo kg/cm 2 (fig. 10). These determi-
nations are in disagreement with the rela-

405 ° C (i.e., well above the high-low inver-
sions) on the high-temperature powder
camera. Quartz was used as the inter-
nal standard. The results were: cristo-
balite, cubic, = 7.1382 ±0.0010 A; tridy-

5000 r

4000

3000

2000

1000

500

1000

Temperature Degrees C

1500

Fig. 10. Curve showing high-quartz-tridymite inversion at pressures up to 1000 kg/cm-

tions obtained from thermodynamic data.
Cell dimensions of tridymite and cristo-
balite (Smith). In view of the inaccuracy
of the previous measurements of cell di-
mensions, Tuttle suggested that they be
measured accurately in order to determine
the pressure dependence of the tridymite-
cristobalite inversion (see fig. 10). Ac-
cordingly, measurements were made at

mite, hexagonal, 0=5.0463 ±0.0020 A, c —
8.2563 ±0.0030 A; volume of cristobalite
cell, 363.72 ±0.2 A 3 ; volume of two tridy-
mite cells, 363.98 ±0.4 A 3 .

At 405 ° C, therefore, the volumes for
the same number of atoms agree within
the experimental error, and, consequently,
the results do not indicate the direction
of the pressure dependence of the inver-

6 2 CARNEGIE INSTITUTION OF WASHINGTON

sion. They do show that the magnitude ene and quartz. This behavior indicated

must be small. It is possible that the cell that the amphibole was unstable in the

volumes change differentially with tern- presence of an excess of water vapor, or

perature, but no measurements have yet that pure magnesian anthophyllite was un-

been made. stable and small amounts of cations other

This remarkable closeness of the cell than magnesium were necessary to stabil-

volumes does not seem so surprising when ize the structure. Experiments have been

the crystal structures are examined, for carried out which indicate that the first

these differ only in the long-range struc- suggestion is more likely correct. An-

ture. thophyllite has now been synthesized by

]adeite (Yoder, Chesterman, Switzer). heating a charge of MgO, Si0 2 , and H 2
The considerable interest in jadeite con- in a sealed platinum tube at about 815 C
tinues. It is still regarded by some as an with 1000 atm external pressure. Tremo-
important constituent of the subcrustal ma- lite has been synthesized by heating mix-
terial, and it figures prominently in the tures of MgO, CaC0 3 , and silica glass of
problem of the so-called high-pressure min- the appropriate composition at 400° C and
erals. Recent reports of the synthesis of 1000 atm water vapor pressure. At 500 C
jadeite by other laboratories have been in- and the same pressure, the mixture crystal-
vestigated but not substantiated. The pres- lizes into talc and diopside. Riebeckite has
ent studies consist in a detailed examina- been synthesized by England, using the
tion of the material collected from the sealed-tube technique. The amphibole was
recently discovered jadeite deposit of San obtained in runs in which less than 3.9
Benito County, California. Chemical anal- per cent water was present with the tem-
yses of four jadeites having different phys- perature below 61 o° C. If the water con-
ical properties have been obtained. One tent and temperature were above these re-
sample by chance has proved to be the spective values, the product was acmite. It
purest jadeite on record. The analyzed has often been observed that the quartz-
materials will be used in the construction perthite granites have an amphibole as the
of a compositional determinative chart principal dark mineral and that the two-
based on X-ray diffraction data and optical feldspar granites carry a mica. This obser-
properties. This investigation is a co-oper- vation supports the contention that the
ative project with Mr. C. W. Chesterman, character and amount of unmixing in the
Division of Mines, State of California, and feldspars are controlled in part by the
Dr. George Switzer, of the U. S. National amount of water vapor. When the amount
Museum. of water present is in excess (indicated by

Stability relations of the amphiboles a mica), the feldspar unmixes completely

(Tuttle, England). The stability relations into two feldspars. When the amount of

of the amphiboles have long been major water present is small (indicated by an

mineralogical problems. Their experi- amphibole), the unmixing of the feldspar

mental investigation has been tedious and is incomplete and perthite results,

the data are few. Anthophyllite was pre- The systems Na 2 — FeO — Si0 2 and

viously synthesized by Bowen and Tuttle Na 2 — MgO — Si0 2 (Schairer, Yoder,

(1949, No. 1111), by decomposing talc in Keene). Data on the stability relations

the presence of water vapor, but continued of the dry silicates at normal atmospheric

heating under the same conditions caused pressure are the cornerstone on which

the amphibole to break down into a pyrox- progress on the more difficult hydrous

GEOPHYSICAL LABORATORY

systems and systems at high hydrostatic tain alkaline rocks and minor mineral
pressures is based. Although studies of constituents of many of the igneous rocks,
dry silicate systems have been in progress During the past year, attention has been
at this Laboratory for the past forty-seven concentrated (Schairer, Yoder, Keene) on
years, much remains to be learned about the ternary system Na 2 — MgO — Si0 2 .
the fundamental relations between the im- This system has proved to be unusually
portant rock-forming oxides. Even though complicated and interesting. So far we have
we have been able to attack the problem recognized eight different ternary com-
of the melting and crystallization relations pounds, all of which have incongruent
in complex quaternary and even quinary melting points and some of which have
systems by separating them into a series very restricted fields of stability on the
of smaller parts and by treating each of liquidus surface. We have prepared 187
these parts as a unit, there still remain a separate ternary compositions and studied
few fundamental ternary systems on which them by the method of quenching. Prep-
data are meager or entirely lacking. aration and study of such a large number

No reliable data are available on the sys- of compositions were required to locate
tern Na 2 — FeO — Si0 2 . Last year a re- accurately the phase boundaries of the
connaissance of this system was begun many fields and the large number of ter-
(Schairer, Thwaite, Yoder) which showed nary invariant points, most of which are
it to be unusually complex and difficult ex- reaction points because of the incongruent
perimentally. The melts were viscous and nature of the ternary compounds' melting
failed to come to equilibrium promptly in behavior. Several of the new compounds
iron crucibles. Many new crystalline phases are so similar in optical properties that
were encountered whose X-ray patterns they are difficult to identify under the mi-
were obtained but whose compositions croscope, particularly when they are pres-
were unknown, and it was evident that ent as tiny crystals suitable for rapid at-
some of them were metastable phases and tainment of equilibrium between crystals
that equilibrium had not been attained, and liquid. Each of the compounds, how-
At this stage it became obvious that further ever, gives a very characteristic X-ray pow-
progress in the system Na 2 — FeO — Si0 2 der pattern by which it may readily be
required a knowledge of the comparable identified. Fortunately, equilibrium be-
system Na 2 — MgO — Si0 2 . A search of tween crystals and liquid is usually at-
the literature revealed that this system had tained in the system in a few hours or
been studied by Russian investigators, but days, depending on the composition. It
a laboratory reconnaissance showed that was necessary to use crystalline material
the Russian data were inadequate and un- for the quenching runs, and even then, in
reliable. some cases where crystallization had not

A knowledge of the crystalline phases taken place in the proper temperature

and melting relations in the two ternary range, a metastable liquidus was deter-

systems Na 2 — MgO — Si0 2 and Na 2 — mined before the stable crystalline phase

FeO — Si0 2 is prerequisite to studies of was encountered.

more complex hydrous systems. The lat- We present here a preliminary phase

ter studies would elucidate the complicated diagram for the system Na 2 — MgO —

compositions of, and relations between, the Si0 2 (fig. 11). Only the fields of the

so-called alkaline pyroxenes and amphi- several primary phases and the ternary

boles, which are major constituents of cer- invariant points are shown. The tempera-

6 4

CARNEGIE INSTITUTION OF WASHINGTON

tures placed at the ternary invariant points in progress. This preliminary diagram is

are only approximate, and their more pre- given at this time so that it may be useful

cise determination is now in progress. The immediately to workers in ceramics and

tie lines are not shown because the compo- metallurgy who require only a general

sition of some of the ternary compounds knowledge of melting temperatures and

/695+5°i

Si0 2

/695±5
70,

/543±2

/557±2°

MgOSi0 2

2MgOSi0 2

/890±40 a /
40.

q 2 0-Si0 2

Na 2
IA/T. PER CENT

Fig. ii. Preliminary phase diagram for the system Na 2 — MgO — Si0 2 , showing fields of primary
phases and the ternary invariant points (equilateral triangle truncated at about 25 per cent Si0 2 ).

is not yet known with certainty. Experi- the compositions and approximate temper-
ments to determine the exact composition atures of the low melting eutectics in this
of each of the ternary compounds are now system.

MICA GROUP

Phase relations in the micas (Yoder, K 2 (Mg 5 Al) (Al 3 Si 5 )0 2 o(OH)4; leucophyl-

Eugster). The following end members of lite, K(MgAl)Si 4 Oi (OH) 2 ; annite, KFe 3 -

the mica group were investigated: mus- (AlSi 3 )Oio(OH) 2 ; paragonite, NaAl 2 -

covite, KAl 2 (AlSi 3 )Oio(OH) 2 ; phlogo- (AlSi 3 )Oio(OH) 2 ; and polylithionite,

pite, KMg 3 (AlSi 3 )Oi (OH) 2 ; eastonite, K(Li 2 Al)Si 4 Oi (OH) 2 . To determine

GEOPHYSICAL LABORATORY

their mutual relations and their stability
relative to those minerals commonly asso-
ciated with them, the following joins were
studied : muscovite — phlogopite, phlogo-
pite — eastonite, muscovite — leucophyllite,
muscovite — quartz, muscovite — clino-

chlore, phlogopite — annite, muscovite — pa-
ragonite, muscovite — montmorillonite.

The unit of structure of mica consists of
two sheets of silica tetrahedra bound by oc-
tahedrally co-ordinated aluminum atoms.
These double sheets are held together by
large potassium (or sodium) atoms in 12-
fold co-ordination. The hydroxyl groups
occupy sites at corners of the aluminum
octahedra. There are four tetrahedral,
three octahedral, and one dodecahedral
sites per unit. The mica structures are
built of one, two, three, or six of these
units. The stacking of these units may be
accomplished in different ways, and it is
this property which permits polymorphism
in the micas.

There are essentially two basic types of
mica structure: one in which all the octa-
hedral positions are filled, as in phlogopite,
and one in which only two-thirds of the
octahedral positions are filled, as in mus-
covite. Complete solid solution between
these two types is not believed possible.
For example, the partial atomic substitu-
tion along the join muscovite — phlogopite
is according to the scheme 2AF 1 ^ 3Mg VI ,
where the roman numerals refer to the co-
ordinated position. In a complete solid-
solution series, such as that along the easto-
nite — phlogopite join, the substitution
wholly within one type is A1 VI A1 IV ^
Mg VI Si Iv . A similar substitution, but re-
taining an unoccupied octahedral site,
takes place along the join muscovite — leu-
cophyllite. A third type of substitution in-
volves the potassium atom. This may go
according to the scheme K XII A1 IV ^ Si IV *
H 2 O xn , as in the soil micas. The substitu-
tions in lepidolite are: Li VI Li VI Si IV ^

A1 VI A1 IV as in the polylithionites, and
Li VI Al IV <=^ Si IV as in the paucilithionites.
Neither of these schemes permits a com-
plete series of solid solutions between mus-
covite and lepidolite. This finding is sup-
ported by recent studies of natural lepid-
olites in the Mineralogical Laboratory of
the University of Michigan. In those sub-
stitution schemes involving Mg +2 , the atom
Fe +2 may proxy. Similarly, sodium may
proxy for potassium as in paragonite.

The chief difficulties in the experiments
on the mica group were slow reaction
rates, change of composition during the
run, difficulty in identifying products, and
formation of metastable products. The
first difficulty was obviated through the
construction of a sufficient amount of ade-
quate hydrothermal equipment to permit
continuous runs up to three months' dura-
tion. The changes of composition during
the run were essentially eliminated by the
use of sealed platinum tubes. The tech-
nique of welding small tubes containing
the sample and water has been perfected
to a fast, routine operation. The fine-
grained nature and imperfect crystalliza-
tion of the products made identification by
the customary optical methods exception-
ally difficult or impossible. X-ray diffrac-
tion and, on occasion, electron diffraction
techniques were relied on for determina-
tion of the phases. Metastable products
were found to persist only in specific mix-
tures of a particular composition. By sys-
tematically preparing all possible combina-
tions of natural and synthetic compounds
in the requisite proportions for each de-
sired composition, mixtures were found
suitable for detailed hydrothermal study.

Muscovite: The syntheses of randomly
stacked one-layer monoclinic muscovite
(iMd), one-layer monoclinic muscovite
(iM), and two-layer monoclinic muscovite
(2M) have been accomplished. The re-
maining known polymorph of muscovite,

56 CARNEGIE INSTITUTION OF WASHINGTON

three-layer rhombohedral (3R), was ob- psi. Above this curve, sanidine + corun-

tained in only one run with 2M muscovite, dum + vapor are the stable phases. The

and its synthesis is therefore not substan- geological implications of this curve are

tiated. In those previous attempts to syn- many. For example, a comparison of this

thesize muscovite for which X-ray data upper stability curve of muscovite with the

have been published, the phase produced minimum melting curve of the granite

is identified by this Laboratory as iM mus- system determined by Bowen and Tuttle

covite. It is believed, therefore, that the suggests that muscovite forms in granitic

present work includes the first synthesis of magmas above approximately 25,000 psi

the common hydroxyl-bearing 2M musco- water vapor pressure, and in the solid state

vite, the mica now used extensively in below that pressure in granitic rocks,

industry. These relations may account for the ap-

The stability ranges of the muscovite parent two generations of muscovite in
polymorphs named could not be fixed ac- some granites. As a corollary to this rela-
curately because of the sluggish nature of tion, muscovite should not be found in one-
the transformations. The following trans- feldspar granites, and preliminary studies
formations, however, were effected : iMd— » of such granites are in accord with this
iM-^2M. The first transformation is prob- conclusion. In addition, the upper stabil-
ably dependent on factors affecting reac- ity curve of muscovite marks the condi-
tion rates. The second transformation may tions of the second sillimanite (or ortho-
be related to a univariant curve of equilib- clase) isograd. This isograd usually marks
rium (i.e., an isograd) . A natural 3R mus- the highest temperatures reached in meta-
covite was converted to 2M after 3 months' morphism, and, therefore, outlines the re-
hydrothermal treatment. gion of experimental conditions in which

The randomly stacked one-layer mono- to investigate the critical isograds at lower

clinic muscovite gives an X-ray pattern temperatures. The sequence iMd— >iM— »

identical with that of material called illite, 2M probably obtains in the metamorphism

a common mineral found in soils. Illite of sediments.

has been loosely defined as a mica contain- Phlogopite: The one-layer monoclinic

ing less potassium and more water than (iM) or the three-layer rhombohedral

muscovite. It is proposed to redefine illite (3R) phlogopite has been synthesized,

as a mica of muscovite composition having These forms are very difficult to distin-

a randomly stacked one-layer monoclinic guish by means of powder X-ray diffrac-

structure. Those micas which lie on the tion patterns, but are easily distinguished

join illite — montmorillonite, having less from the 2M phlogopite, which has not

potassium and more water than muscovite, been synthesized. A curve through the

may be called hydromicas. Illite, so de- points 840° C and 1000 psi water vapor

fined (iMd), can easily be distinguished pressure, 995 ° and 5000 psi, 1025 and

from iM muscovite by the absence of the 10,000 psi, and 1050 ° and 15,000 psi marks

h\l reflections in the powder X-ray difrrac- the upper limit of stability of phlogopite.

tion pattern. Above this curve, the stable phases are

The upper stability limits of muscovite forsterite + leucite + kalsilite + vapor,

have been determined and are represented These same phases form metastably at

by a curve passing through the points temperatures immediately below the curve,

630° C and 5000 psi water vapor pressure, particularly at low pressures.
665 ° and 15,000 psi, and 705° and 30,000 The phlogopite forms quickly at the

GEOPHYSICAL LABORATORY

6 7

higher temperatures and grows very slowly eastonite — annite — siderophyllite. Recon-

at the lower temperatures. It has not been naissance studies on the joins phlogopite —

ascertained whether the polymorphs are eastonite and phlogopite — annite show that

related by a univariant curve of equilib- the biotites can be synthesized. Although

rium or are dependent on growth factors, their properties and relations have not been

The phlogopites from lavas are one-layer completely established, there is a complete

monoclinic (iM) ; iM phlogopites are also series of solid solutions between phlogopite

found, however, in metamorphosed lime- and eastonite. The members of the series

stones. In the same hand specimen, two or are determined by the change in unit cell

more of the polymorphs may be found, size.

No clear relation between the temperature Experiments on reactions alleged to rep-

of formation and the type of layering in resent the biotite isograd indicate that

natural phlogopites has yet been deter- some muscovite-chlorite-quartz schists may

mined. undergo the following reaction: musco-

Phlogopite appears on the liquidus in vite + 2 clinochlore + 4 quartz <=^ phlog-

the synthetic "haplogranite" under hydro- opite + 2 cordierite + 8 water. It can be

thermal conditions. The formation of concluded that if this reaction takes place,

phlogopite (and biotite) and the absence the growth of biotite in low-grade meta-

of muscovite in lavas are accounted for by morphic rocks must be accompanied by

their stability curves. another phase, such as cordierite. It has

Phlogopite-museovite : Besides the im- been previously shown that the composi-

portant problem of polymorphism in the tion of natural biotites does not lie on the

micas, there is the question of the nature join connecting the compositions of natu-

and extent of solid solution between the ral chlorites and natural muscovites. This

various mica end members. Most signifi- strongly supports the view that biotite is

cant of these is the common association of but one member of a reaction pair. It has

a phlogopitic mica (i.e., a biotite) and mus- been observed that cordierite or andalusite

covite. All field evidence points to limited accompanies the formation of biotite. On

solid solution between muscovite and the assumption that the parent rock under-

phlogopite, with one exception, the exist- going metamorphism is dominantly a

ence of a mineral called mahadevite alleged muscovite-chlorite-quartz schist, the ap-

to lie midway in composition between pearance of cordierite or andalusite may be

muscovite and phlogopite. A recast of the regarded as an indication that the condi-

four published analyses of mahadevite tions of the biotite isograd have been

shows, however, that these minerals actu- reached. Whether biotite, cordierite, or

ally lie on the join eastonite — muscovite andalusite is most conspicuous in the rock

and are perhaps the purest eastonites would depend on the bulk composition of

known. The experimental studies indicate the parent schist and probably also on

that there is solid solution between musco- growth factors. It is interesting to note

vite and phlogopite. The extent of the that the above reaction has been carried

solid solution is being determined by out in the laboratory at temperatures as

means of the changes in the unit cell di- low as 300° C.

mensions as measured by X-ray diffraction. Seriate: The very fine-grained micas

Biotite: The common mica biotite is common in the schists are loosely lumped

complex in composition, but can be closely under the term sericite. Most such micas

represented by the system phlogopite — are normal muscovite, but some have been

CARNEGIE INSTITUTION OF WASHINGTON

identified as high-silica mica (phengite)
and paragonite, the soda mica. For this
reason the muscovite — leucophyllite, mus-
covite — quartz, muscovite — paragonite, and
muscovite — montmorillonite joins were in-
vestigated. Both leucophyllite and mont-
morillonite have more silica than musco-
vite. In addition, it is conceivable that the
fine-grained micas may contain small
amounts of lithium, which would lead to
a higher silica content if the structural
scheme indicated for polylithionite ob-
tained.

Present data indicate that the primary
mica formed from compositions on the
muscovite — montmorillonite join is a hy-
dromica as defined above. The phases
formed from compositions on the quartz —
muscovite join were quartz + hydromica
+ sanidine. There was not sufficient evi-
dence to determine whether or not solid
solution was taking place according to the
postulated structural scheme on the join
muscovite — leucophyllite. Paragonite was
synthesized, but no definite conclusions
concerning the nature of the solid solu-
tions with muscovite have yet been estab-
lished. The synthesis of hydroxyl-bearing
polylithionite was not accomplished. This
failure may be due to the fact that most
natural lepidolites contain considerable flu-
orine instead of hydroxyl. Although it is
not known whether or not fluorine is a re-
quired constituent of the lepidolites, a poly-
lithionite with some fluorine has been syn-
thesized.

In order to cast some light on the nature
of the very fine-grained micas in sedi-
ments, a new approach was employed. In
the weathering and transport of rocks, the
micas undergo severe treatment. One of
the processes which reduce the mica is
abrasion. A normal 2M muscovite was
ground dry in an agate mortar for 30
hours and samples were taken at 2-hour
intervals. At the end of 6 hours of dry

grinding, no crystal structure could be
found from X-ray diffraction. At the end
of 30 hours, no crystal structure could be
found from electron diffraction, a tech-
nique capable of 5 A resolution. The mica
had apparently been reduced to or below
its fundamental unit dimensions, approxi-
mately 5X9X10 A. On recrystallization
under hydrothermal conditions this mate-
rial behaved in the same manner as other
preparations, transforming in sequence
from iMd to iM to 2M. It is possible that
some sediments contain such unidentifi-
able material which when metamorphosed
produces mica.

The mica polymorphs (Smith, Yoder).
A theoretical study of the ways of stacking
mica layers showed that there are only five
simple ways of doing this which produce
regular ordered structures: 1 -layer mono-
clinic, 2-layer monoclinic, 3-layer trigonal,
6-layer monoclinic, and 6-layer hexagonal.
Disordered structures may be built which
have randomly arranged slabs of layers.
If only a very occasional stacking fault oc-
curs, twinned crystals are produced. The
possible twin laws were deduced from the
theoretical study. A very simple way of
describing the stacking was developed
which consists simply in showing the posi-
tions of the potassium atoms, thus reduc-
ing the structure to its elementary simplic-
ity and enabling an easy understanding of
the various types of stacking.

Six structures have been observed experi-
mentally by Hendricks and Jefferson in
their classic study of polymorphism in the
micas. One of these, their 6-layer triclinic,
has been shown by experimental work to
be a differently described 2-layer mono-
clinic form. Their 24-layer triclinic crys-
tal has not been investigated further, but
there is a suspicion that it also is 2-layer
monoclinic. The other four structures are
identical with the first four found theo-

GEOPHYSICAL LABORATORY

retically. The remaining theoretical struc-
ture, the 6-layer hexagonal, has not been
found experimentally. The twinned crys-
tals observed confirm one of the theoretical
twin laws; the morphological studies de-
scribed in the literature confirm another.

The agreement between theory and ex-
periment is therefore very good. Accurate
measurements on the precession camera of
i-layer monoclinic, 2-layer monoclinic, and
3-layer trigonal crystals show that the
assigned symmetry is exact and not merely
pseudosymmetry. This conclusion answers
various speculations presented in the liter-
ature.

The more important problem of dis-

covering the factors which govern the
growth of a particular polymorph in a
certain set of geological conditions has not
yet been solved. For example, muscovite
prefers the 2-layer monoclinic structure,
whereas phlogopite prefers the 1 -layer
monoclinic or 3-layer trigonal structure.
Unfortunately, no accurate structural anal-
ysis exists, and we know only that the
layers in muscovite are buckled whereas
the layers in phlogopite are planar. The
answer to the problem undoubtedly lies
in accurately determining the atomic po-
sitions in the polymorphs and then work-
ing out the structural forces from those
positions.

OLIVINE AND ZEOLITE GROUPS

Olivine (Yoder) . The study of the com-
position of natural and synthetic olivine
by X-ray methods has been greatly en-
larged. Twelve synthetic and 23 analyzed
natural olivines have been investigated, in
addition to a host of natural olivines whose
compositions were determined by optical
methods. With the exception of two natu-
ral specimens, the composition of olivine
can be estimated to ±2 mole per cent by
means of the X-ray method. The details
of the technique and the determinative
chart are being prepared for publication.

High-pressure behavior of analcite (Yo-
der, Weir). Compressions of analcite, a
mineral found in rocks of the type believed
to exist deep in the earth, indicate that a
polymorphic transition takes place at ap-
proximately 8400 atm (equivalent to 34
km depth) and room temperature. The

high-pressure transition was deduced from
a small, but distinct, discontinuity in the
volume decrements of analcites from
Golden, Colorado, and Wassons Bluff,
Nova Scotia. Two samples from the Tyrol,
Austria, cut from some of the largest
known single crystals, did not show a dis-
continuity in the volume decrements, but
gave an abnormal increase of compressi-
bility with pressure. Such behavior usually
precedes a transition, and it is predicted
that the transition in the Austrian samples
lies at pressures slightly higher than 10,000
atm, the limit of the present experiments.
There is no doubt, therefore, that a high-
pressure modification of at least one rock-
forming silicate does exist. This study
was done in collaboration with Mr. Charles
E. Weir, of the National Bureau of Stand-
ards.

CONTRIBUTIONS OF THERMOCHEMICAL AND X-RAY DATA
TO THE PROBLEM OF MINERAL STABILITY

T hermochemical properties of minerals calorimeter has been brought to a stage

(Kracek, Neuvonen, Burley, Gordon), where it is possible to evaluate the heats

During the past year the thermochemical of formation of many silicates, especially

work with the hydrofluoric acid solution those containing alkali and aluminum

7 o

CARNEGIE INSTITUTION OF WASHINGTON

atoms. The heats of formation are funda-
mental physical constants which form the
basis for theoretical deductions from ex-
perimental thermochemical results. The
determination of heats of formation of sili-
cates containing alkali and aluminum has
not been possible heretofore because reli-
able heats of solution of the alkali oxides
and of alumina as corundum were not
available. The heats of solution of these
oxides must be measured under the same
experimental conditions that prevail in the
determination of the heats of solution of
the minerals concerned; the experimental
difficulties which prevented earlier deter-
mination are twofold: (a) the uncertain-
ties involved in preparing and handling as
well as dissolving reproducibly the alkali
oxides in relatively strong acid at the tem-
perature at which our calorimeter is oper-
ated, and (b) the insolubility of corundum
in the 20 per cent hydrofluoric acid em-
ployed in the calorimeter. There are sev-
eral types of aluminum oxide which are
soluble in the acid, but these are in general
somewhat indefinite products which would
not serve as standard materials to be used
in evaluating the heats of formation. The
afore-mentioned difficulties were avoided
by the methods to be briefly described.

1. The alkali oxides: The heats of solu-
tion of lithium, sodium, and potassium
oxides in 20 per cent hydrofluoric acid at
74.7 ° C were determined from two series
of thermochemical measurements based on
the heat of solution of the alkali chlorides
and the heat of solution of the alkali sul-
fates, together with the heats of other ap-
propriate reactions.

The chloride scheme of reactions in-
volved: (1) heat of solution of the given
alkali chloride, (2) heat of formation of
the alkali chloride from the elements, (3)
heat of formation of gaseous hydrogen
chloride from the elements, (4) heat of
solution of hydrogen chloride in water to

form concentrated hydrochloric acid, (5)
heat of solution of concentrated hydro-
chloric acid in the calorimetric acid at
74.7 , (6) heat of formation of liquid
water, (7) heat of formation of the given
alkali oxide, and (8) heat of solution of
the proper quantity of water at 25° in the
calorimetric solution at 74.7°.

The sulfate scheme of reactions involved
a parallel series of reactions, with the sub-
stitution of alkali sulfate for alkali chlo-
ride, and of sulfuric acid for hydrochloric
acid.

In both cases the heats of reactions (1),
(5), and (8) were measured, and those of
reactions (2), (3), (4), (6), and (7) were
taken from the literature. The values ob-
tained by the two schemes and their re-
conciliation are presented in table 3.

TABLE 3

Heats of solution of alkali oxides in 20 per
cent hydrofluoric acid at 74-7° c

Oxide

Reaction scheme

Average*

Differ-

Chloride*

Sulfate*

CI - SO4

Li 2 0...
Na 2 0. .
K 2 0. ..

-59.36

-87.66

-109.61

-60.86

-89.19

-110.37

-60.11

-88.42

-109.99

1.50

1.53
0.76

* AH in kcal/mole oxide.

2. Corundum: The heat of solution of
corundum, the stable form of alumina, in
20 per cent hydrofluoric acid at 74.7° C
was determined from the following series
of reactions: (1) heat of solution of pure
aluminum, (2) heat of solution of an acid-
soluble form of alumina, (3) heat of for-
mation of liquid water, (4) energy to
heat gaseous hydrogen from 25° to 74.7°,
(5) heat of solution of liquid water at 25 °
in the calorimetric solution at 74.7°, (6)
heat of formation of the particular acid-
soluble alumina, obtained as the resultant
of the heats of reactions (1) to (5), (7)
heat of formation of corundum, (8) heat

GEOPHYSICAL LABORATORY

of transition of the soluble alumina to
corundum, obtained as the resultant of
reactions (6) and (7), (9) heat of solution
of corundum in the calorimetric solution,
obtained as the resultant of reactions (2)
and (8).

Several acid-soluble aluminas were pre-
pared for the purpose of measuring the
heat of reaction (2). They were made
from reagent-grade aluminum hydroxide
powder by slowly raising the temperature
to 350° C and holding at this temperature
for about 24 hours. The composition of
the resulting product was approximately
A1 2 3 '0'5H 2 0. The temperature was
then slowly raised to 600 ° and maintained
for 3 or more days. This treatment re-
sulted in a product containing in all cases
less than 1 per cent water. The X-ray pat-
tern agreed with that of gamma alumina,
as defined by the Research Laboratory of
the Aluminum Company of America. The
particular product to which all the heats
of formation will be referred was prepared
from Baker's Analyzed C. P. Aluminum
Hydroxide Powder, AI2O3 + *H 2 0, lot
1 1 1428, for which x was found to be 3.019
moles. The lot analysis indicated, "Sodium
a trace." U. S. Geological Survey flame
photometer analysis gave 0.42 per cent
Na 2 and 0.04 per cent K2O. Correction
was made for these on the assumption that
they were combined in the preparation as
alkali aluminate; the heat of solution of
NaA10 2 was measured on a sample pre-
pared by Schairer. In making the correc-
tion, the small amount of potassium oxide
was included with the sodium in the
sodium aluminate. Other impurities were
of negligible amount.

The pure aluminum dissolved in reac-
tion (1) was furnished by the Research
Laboratory of the Aluminum Company of
America. It contained as impurities: cop-
per, 0.000; iron, 0.001; silicon, 0.002; mag-
nesium, 0.001 per cent. Aluminum by dif-

ference was 99.996 per cent. No correc-
tion was applied for the small amounts of
the impurities; a relatively large correction
was necessary, however, for the heat lost
by evaporation of water and hydrogen
fluoride from the calorimetric solution by
the evolved hydrogen. The aluminum was
cut in strips, etched with hydrofluoric acid
to remove surface iron, dried with alcohol
and ether which were evaporated in an
oven for a short time, and used as soon
thereafter as possible.

Numerical data: 1. Alkali oxides: The
heat of solution of the alkali oxides may be
expressed by the following type of reac-
tion: Me 2 0(c, 25° C) + 2 H + (sol, t°) =
2 Me + (sol, t°) + H 2 0(sol, t°) AH = x,
where Me denotes Li, Na, or K; c, crystal-
line; sol, 20 per cent hydrofluoric acid; t°,
74.7 ° C; AH, heat of reaction per mole of
oxide; x, numerical value of the heat of
reaction AH.

2. Corundum: The solution reaction in
this case is: Al 2 3 (c, 25 ° C) + 6H +
(sol, t°) = 2A1 +++ (sol, t°) + 3 H 2
(sol, t°) AH = x, where the symbols have
the same significance as above. In view
of the importance of this reaction, a de-
tailed summary of the heats of reaction in-
volved in this determination is given in
table 4. Among the heats of reaction in
this table, those for reactions (1), (2), and
(5) were measured, those for (3), (4), and
(7) were taken from the literature, and
the remaining ones, (6), (8), and (9),
were computed from the others. We may
call attention to the small heat of transi-
tion for reaction (8), which is smaller
than has been estimated by others, and
smaller than its uncertainty.

Heats of formation of silicates: Meas-
urements were carried out of the heats of
solution of a number of silicate minerals,
some of which have already been pub-
lished, and of several alkali silicates of in-
terest for the work of this Laboratory and

CARNEGIE INSTITUTION OF WASHINGTON

for petrological research in general. For
computing the heats of formation, the
needed heats of solution of the component
oxides have been measured or evaluated as
indicated above. The heats of formation
are referred to the oxides, and in particular

TABLE 4

Heats of solution of corundum in 20 per cent
hydrofluoric acid at 74-7° c

Reaction AH (cal)

(1) Heat of solution of 2A1 -287,590±420

(2) Heat of solution of soluble

alumina -92,136±410

(3) Heat of formation of 3H 2

at 25° C -204,951±30

(4) Heat to raise 3H 2 from 25°

to 74.7° C 1,036±20

(5) Heat of solution of 3H 2 at

25° in (sol, t°) 2,364±15

(6) Heat of formation of soluble

alumina from the elements

at 25° -399,077±590

(7) Heat of formation of corun-

dum from the elements at

25° -399,090±240

(8) Heat of transition of soluble

alumina to corundum at

25° -13±640

(9) Heat of solution of corun-

dum at 25° in (sol, t°) . . . -92,123±760

to silica as quartz, and alumina as corun-
dum. The heats of solution are corrected
so that the resultants of the thermochem-
ical summations are heats of formation at
25 ° C = 298.1 6° K. This may be repre-
sented technically by the process : materials
at 25° C dissolved at t° = 74-7° C. The
alkali silicates were prepared in the ther-
mochemical laboratory; the other silicates
are natural or synthetic materials, for most
of which the heats of solution have already
been published. The values are given in
tables 5 to 8. Table 5 gives the heats of
solution of the oxides; table 6, the heats
of solution and formation of the alkali
silicates, together with the heats of phase
change at 25° C; table 7, the differences in
the heats of solution for homologous series
of silicates of a given alkali metal in com-
parison with the heats of solution of the
various forms of silica; and finally, table 8,
the heats of formation for the minerals
studied.

Significance of the silicon-oxygen and
aluminum-oxygen distances in paracelsian
and high-came gieite (Smith). Several
minerals (including the feldspars) have
both aluminum and silicon in tetrahedral
co-ordination. The structural changes in

TABLE 5
Heats of solution of certain oxides, dissolved from 25 C in 20 per cent

HYDROFLUORIC ACID AT 74-7° C

Oxide How measured

Li 2 Indirectly, average value

Na 2 Indirectly, average value

K 2 Indirectly, average value

MgO Directly, by Sahama and Torgeson

CaO Directly

AUO3 (soluble) Indirectly

A1 2 3 (corundum) Indirectly

Si0 2 (quartz) Directly

Si0 2 (cristobalite) Directly

Si0 2 (glass) Directly

Ga 2 3 (beta)* Directly

* Supplied by J. R. Goldsmith, University of Chicago.

tf(kcal/mole)

-60.11

±0.75

-88.42

±0.76

-109.99

±0.38

-38.75

-55.72

±0.07

-92.136±0.410

-92.123±0.760

-33.30

±0.04

-33.93

±0.03

-35.48

±0.02

-51.61

±0.75

GEOPHYSICAL LABORATORY

TABLE 6
Thermochemical properties of certain lithium, sodium, and potassium silicates

Substance and condition

AH (sol, 74.7° C)
(kcal/mole)

Latent heats of phase
change at 25° C, x ->- gl

(kcal/mole)

(cal/g)

AH f at 25° C
from oxides
(kcal/mole)

Li 2 Si0 3 (crystalline) -58.96

Li 2 Si 2 5 (glass) - 103 . 56

Li 2 Si 2 0s (crystalline) — 92 . 13

Na 2 Si03 (crystalline) — 65 . 66

Na 2 Si 2 5 (glass) -104.70

Na 2 Si 2 5 (E) - 101 . 98

Na 2 Si 2 5 (A) -100.04

Na 2 Si 2 5 (B) -99.45

K 2 Si 2 5 (glass) - 103 . 54

K 2 Si 2 5 (crystalline) -99.07

K 2 Si 4 9 (glass) -173.87

K 2 Si 4 9 (crystalline) - 169 . 25

x, crystalline; gl, glass; E, A, B, polymorphous forms.

11.42

2.72
4.66
5.25

4.47
4.62

76.2

14.9

25.5
28.8

20.9
13.8

-34.45
-23.15
-34.58
-56.06
-50.32
-53.04
-54.98
-55.57
-73.05
-77.52
-69.32
-73.94

TABLE 7

Differences for i Si0 2 increment in the heats of solution of alkali silicates

(in kilocalories per mole)

Silicate Difference in heat of solution

Lithium (disilicate-metasilicate) —92.13+58.96 = —33.17

Sodium (disilicate E-metasilicate) —101.98+65.66 = —36.32

Sodium (disilicate A-metasilicate) —100.04+65.66 = —34.38

Sodium (disilicate B-metasilicate) —99.45+65.66 = —33.79

Potassium (tetrasilicate-disilicate) ^(-169.25+99.07) = -35.09

For comparison :

Si0 2 (quartz) -33 . 30

Si0 2 (cristobalite) -33 . 93

Si0 2 (glass) -35 .48

the feldspars, such as the change caused by
sanidinization of orthoclase, are almost
certainly caused by order-disorder of the
silicon and aluminum atoms. Since the
electron densities of silicon and aluminum
are too close for distinction by X-rays, we
have to rely on measurements of Si-O and
Al-O distances, and it is very important
that these be known accurately. Although
good values have been obtained for Si-O,
giving Si-O = i.6o±o.oi A, no suffi-
ciently accurate measurements have pre-

viously been made for Al-O. In paracel-
sian, Sii/ 2 Ali/ 2 -0 is 1.696 ±0.01 A; and in
carnegieite Si-O = 1.587 ±0.02 A, and
Al-O = 1.767 ±0.02 A. The reliable values
of (Si, Al-O) are plotted against percent-
age of Al in figure 12. It will be seen that
the measured values are consistent with a
linear variation between Si-O = 1.60 ±
0.01 A and Al-O = 1.77 ±0.02 A. The
result was encouraging, for it was previ-
ously thought from a consideration of the
older, less accurate values that Si-O = 1.62

CARNEGIE INSTITUTION OF WASHINGTON

TABLE 8
Thermochemical properties of certain silicate minerals

AH/ at 25° C
from oxides
(kcal/mole)

Mineral (sol, 74.7° C)

(kcal/mole)

Albite, natural, Amelia, Virginia — 149 . 79

Albite, natural, Varutrask, Sweden — 148 . 12

CaAl 2 Si 2 08, anorthite, synthetic — 185 . 11

CaAlGaSi 2 8 , Ga anorthite, synthetic* — 170 . 80

Orthoclasef — 144 . 54

Nepheline, synthetic — 87 . 33

Jadeite, Burma — 1 14 . 96

Jadeite, Japan — 115 . 14

Leucite, Vesuvius — 1 16 . 08

Diopside, synthetic —123.31

Wollastonite, synthetic —65.36

Pseudo-wollastonite, synthetic —66.92

* Supplied in the form of glass by J. R. Goldsmith, University of Chicago.

t Value derived from heats of solution of iron-bearing orthoclases from Madagascar.

-40.38
-42.05
-29.33
-23.38
-56.41
-36.24
-41.91
-41.73
-51.57
-37.76
-23.66
-22.10

and Al-O = 1.72, a difference only half
as great as that now found. By careful
work, using two-dimensional analyses, it
should be possible to determine the order-
ing of the aluminum and silicon atoms.

Si-0 AI-0

Fig. 12. Variation of the (Si, Al-O) distance
in oxygen tetrahedra from Si-O to Al-O. The
values for Si-O are for high- and low-quartz,
high- and low-cristobalite, diopside, afwillite,
tilleyite, and high-carnegieite.

Previously it was thought that only the
very laborious three-dimensional analyses
could do this. (It should be noted that
these values are the mean of the four tet-
rahedral distances in the oxygen tetra-
hedron.)

A careful review of the most accurately
measured Si-O distances shows that there
is no evidence that the mean Si-O distance
in any one tetrahedron is affected either by
the method of linkage of the tetrahedra to
form chains, sheets, frameworks, etc., or
by the presence of neighboring cations (ex-
cept possibly by very extreme distribution) .
There is, however, evidence that individual
Si-O distances vary in such a way that an
oxygen atom linked to two silicon atoms
is 0.05 A more distant from the silicon
atoms than an oxygen atom linked to only
one silicon atom.

These results show that to a high degree
of approximation the silicon tetrahedron
contributes about the same amount of lat-
tice energy to each of the various minerals
which contain it, for lattice energy is a
function of interatomic distance.

The crystal structure of paracelsian

GEOPHYSICAL LABORATORY

(Smith). A preliminary report of this de-
termination was given in the last annual
report. The structure analysis was com-
pleted by the aid of a new modification of
the Fourier series, which showed very
clearly the reason for the pseudosymmetry
of paracelsian. It is caused by very small
movements of related atoms which destroy
the orthorhombic symmetry to produce the
lower monoclinic symmetry. These move-
ments are almost certainly produced by
segregation of aluminum and silicon
atoms. It appears likely that when first
formed, paracelsian has true orthorhombic
symmetry, resulting from complete dis-
order of the aluminum and silicon atoms.
On cooling, the thermal motion which pro-
duces the disorder is lessened, and the
aluminum and silicon atoms then segre-
gate into their preferred positions, thus
giving monoclinic symmetry. This be-
havior is the same as that thought to exist
in the feldspar series.

Comparison of the feldspar structure
with the paracelsian structure shows much
similarity. The two structures are built up
of almost identical spirals, the difference
being in their cross-linkage. As this should
provide only a small energy difference, it
undoubtedly is the reason why celsian (a
feldspar) and paracelsian occur together.

Some useful data on the interatomic
distances for barium atoms were obtained
and compared with those in barium titan-
ate, a ferroelectric material with intriguing
commercial possibilities.

The crystal structure of high -came gieite

(Smith). High-carnegieite is the stable
form of NaAlSi0 4 above 1248° C. Below
this temperature nepheline is the stable
form, but high-carnegieite exists below
1248 ° C as a metastable form, and at
690° C there is a high-low inversion to
low-carnegieite. The new determinations
give a more complete explanation of the
X-ray data than did the work in the Lab-
oratory by Barth in 1932. The structure
has the same framework of tetrahedra as
cristobalite except that half the silicon
atoms are replaced by aluminum. To bal-
ance the charge, an equal number of so-
dium atoms occupy holes in the cristobalite
framework. Like cristobalite, carnegieite
has a statistically random arrangement of
oxygen atoms such that they appear to
rotate in a circle of 0.5 A radius. Whether
this is a true relation or only a space and
time average cannot be determined. This
rotation is surely a remarkable phenome-
non in a rigid crystal which has an exact
repeat distance for hundreds of unit cells.
X-ray crystallographers think that many
silicates are disordered at high tempera-
tures, though not always so badly as this.
It is desirable that much work be done on
these high-temperature crystals in order to
elucidate the principles which govern their
structures. All minerals of igneous origin
are formed at high temperatures, and for
an understanding of their growth a study
of high-temperature structures is pertinent.
The high-low inversion of carnegieite is
caused by either (a) collapse of the high
structure around the holes, or (b) cessa-
tion of rotation, or both.

APPLICATION OF TRACE ELEMENTS

Trace elements in some metamorphic
rocks (Eugster). The optical spectrograph
has been a very useful tool in the study
of cosmic and terrestrial abundances of
trace elements. In the past decades a large

fund of information has been accumulated
on average concentrations and modes of
occurrence of most elements; but few at-
tempts have been made to apply this
knowledge to specific geological problems.

7 6

CARNEGIE INSTITUTION OF WASHINGTON

The distribution pattern of certain key nite signs of chemical precipitation, as well

elements will depend largely on the course as biological influence.

of the fractionation processes acting dur- Some of the results of the investigation

ing the formation of the rocks. There is can be summarized as follows:

little doubt that the statistical treatment Alkali metals. For the alkali metals the

of the behavior of trace elements in a par- best correlation was found in the pair po-

ticular geological environment represents tassium-rubidium, as was to be expected.

a powerful tool for a variety of problems. For all inorganic rocks the ratio potas-

In order to test this viewpoint, a suite of sium/rubidium assumes a nearly constant

33 metamorphosed sediments was studied value of about ioo. The potassium/rubid-

with an optical spectrograph. The sped- ium ratio is also constant for all rocks with

mens are from an Alpine area (Aarmas- organic influence such as coal ashes, gra-

siv) on which detailed petrological studies phitic phyllites, and graphite quartzites,

had been made earlier. Such detailed geo- but with a definite enrichment of rubidium

logic information is a prerequisite to intel- over potassium (potassium/rubidium about

ligent sampling. Duplicate measurements 65). The concentrations vary for coal

for 18 elements were carried out at the ashes from 10 to 100 ppm Rb 2 0, and for

Cabot Spectrographic Laboratory of the the other rocks from about 100 to 1800 ppm

Department of Geology and Geophysics, Rb 2 0.

Massachusetts Institute of Technology, Cesium values lie very close to the de-
Cambridge, Massachusetts. The elements tection limit. Cesium seems to behave very
included lithium, sodium, potassium, ru- much like rubidium, although in coal
bidium, cesium, copper, gallium, lead, sil- ashes there is a slight enrichment of cesium
ver, thallium, titanium, zirconium, nickel, over rubidium.

vanadium, scandium, chromium, stron- Sodium increases in coal ashes with in-

tium, and boron. All trace elements pres- creasing total alkali content, but definitely

ent in amounts above the normal detection decreases in rocks without organic influ-

limit were determined for each sample ence, with increasing potassium content, so

without resort to chemical concentration, that for these rocks the sum Na 2 + K 2

The rocks involved in this investigation stays nearly constant,

are probably of Carboniferous age (Ste- Lithium is clearly enriched in hornfelses

phanian) and were deposited in a small in- near the granite contacts, probably by

tracontinental lake during the Hercynian pneumatolytic processes, but otherwise fol-

orogenesis. They suffered an over-all re- lows sodium rather closely,

gional and dynamic metamorphism during The concentration of the alkali metals

the Alpine orogenesis, including strong is probably to a great extent governed by

shearing forces and differential move- differential absorption from true solutions

ments. Furthermore, some hornfelses are by clay particles, although detrital influ-

to be found at the contacts of granites and ences, especially from plant residues, can-

monzonite-batholiths. The main rock not be excluded.

types are now phyllites (chlorite-sericite), Chromium, nickel, scandium, titanium,
biotite-amphibole schists, hornfelses (sil- zirconium, gallium. Addition of very flne-
limanite, chiastolite, or cordierite-garnet) , grained heavy minerals is probably re-
graphitic quartzites, and small coal lenses, sponsible for the presence of these ele-
The sedimentation was to a large extent ments.
mechanical (detrital) but also shows defi- Chromium and nickel show in all in-

GEOPHYSICAL LABORATORY

organic rocks a very good correlation over
a wide range of concentration (5-1000
ppm). Both are probably located in the
same sulfides or silicates; in the silicates,
mainly in augite and amphibole, in which
Cr +3 substituted for Fe +3 , Ni +2 for Mg +2 .
This correlation holds even for most coal
ashes, but with a markedly greater scatter.

Scandium: The correlation between
scandium and nickel or chromium is poor
for all inorganic rocks and very poor for
the coal ashes. This means that most of
the chromium and nickel must be con-
tained in sulfides, because Sc" h3 is typically
lithophile and substitutes for Mg +2 and
Fe +2 in the same silicates as do chromium
and nickel, as part of the complex substitu-
tion Sc +3 A1 +3 <=± Mg +2 Si +4 . This must be
particularly true for the coal ashes.

Titanium shows a behavior rather simi-
lar to that of scandium, especially in the
inorganic rocks. Therefore, it can be sus-
pected that most of the titanium occurs in
the same complex silicates as scandium,
and that only a small or constant amount
of special titanium minerals such as rutile
and sphene is present. This does not hold
to the same extent for coal ashes, where a
greater part of the titanium seems to be
contained in rutile or sphene.

Zirconium increases in coal ashes paral-
lel to scandium and titanium, but charac-
teristically decreases in the same rocks with
increasing chromium and nickel content.
In all the inorganic rocks zirconium shows
a very uniform distribution, with values
between about 140 and 200 ppm. This uni-
formity is probably due to the presence of
a constant amount of zircon.

Gallium, mainly replacing Al +3 , reveals
in these rocks the same trends as in most
other rock types, a very uniform concen-
tration of about 12-30 ppm. Only in a few
coal ashes do the values drop to 3-5 ppm.

Vanadium, copper, boron, strontium.

Vanadium, as would be expected, dis-
closes excellent correlation with scandium
and titanium, as far as the inorganic rocks
are concerned, the vanadium concentration
varying from 100 to 350 ppm. V +4 substi-
tutes for Ti +4 ; V +4 and V +5 substitute for
Al +3 and Fe +3 in the complex silicates. On
the other hand, in most of the coal ashes
a strong enrichment of vanadium can be
found, with values up to 1600 ppm, which
is typical for biological environment. But
some of the coal beds from a slightly dif-
ferent region gave exceptionally low val-
ues (30-50 ppm). Because several coal
beds are of allochthone origin, these low
values may be related to washing out of
the vanadium from the plant residues.

Copper occurs in a completely erratic
way, and no correlation with any of the
other elements (especially chromium and
nickel) could be found. Nevertheless,
most of the coal ashes show a definite en-
richment in copper (25-160 ppm) as com-
pared with the other rocks (6-50 ppm).

Boron concentrations were found to vary
between the limits of 3 and 180 ppm and
to be independent of the distribution of
any other element. High boron contents
are typical for all hornfelses (pneumato-
lytic processes), as well as for some of the
residual schists (detrital tourmaline). The
values for coal ashes are scattered over the
whole range, with biological enrichment
only in certain coal beds.

Strontium values are rather low (30-
200 ppm) in most of the phyllites, espe-
cially in aluminum- and potassium-rich
sericite phyllites, and are significantly
higher (250-980 ppm) in all biotite-amphi-
bole schists. Sr +2 substitutes for Ca +2 . Bio-
logical enrichment must be suspected in
some of the coal ashes (100-270 ppm),
whereas others are extremely poor in
strontium (5-10 ppm).

The remaining elements (lead, silver,

7 8

CARNEGIE INSTITUTION OF WASHINGTON

thallium) could be detected only in certain
specimens, and they occur in too small
amounts to permit any conclusions.

For the inorganic rocks no significant
differences could be found between the
different rock types, for instance phyllites,
schists, hornfelses. Therefore, if migration
took place during the recrystallization, it

was of insignificant magnitude. An excep-
tion is the enrichment of lithium and bo-
ron in hornfelses near the granite contacts.
These few data show that the main
trends of sedimentation for this particular
environment, mechanical, chemical, and
biological, are clearly reflected in the distri-
bution of the trace elements.

AGE OF ROCKS

In 1952, a co-operative program was be-
gun by the Department of Terrestrial Mag-
netism and the Geophysical Laboratory to
determine the age of rocks by use of special
methods on the various minerals in the
rock, and to make a critical comparison
of the reliability of these methods. This
program has been actively followed during
the past year. At the present time, how-
ever, such a critical comparison has not
been completed. Therefore the ages given
are simply the numerical results of the
analyses, using the now accepted half lives
and decay schemes of the elements in-
volved. Any future changes in these may
materially affect the final values.

Isotope dating of igneous intrusives
(Aldrich, Davis, Graham, Nicolaysen, Til-
ton) . There are many reasons for develop-
ing techniques to fix accurately the age of
rocks, but one of the most immediate is
the demand for raw materials which forces
the practical geologist to find new source
areas. The search for such areas is usually
based on the assumption that ore bodies
tend to accumulate in similar geological
environments and at similar geological
times of mineralizing activity. It is pos-
sible for the geologist to locate those sim-
ilar places which could be areas for ore
deposition, but he is not always able to
determine whether or not the constituent
rocks were formed at a time when ore
deposition is known to have taken place.

Accurate age determinations will greatly
facilitate such a search.

There are many methods for determin-
ing the age of a rock, some absolute, others
relative. Until recently, methods of abso-
lute age determination have been limited
in application to ores of radioactive ele-
ments (e.g., uranium and thorium) for
which chemical determination of the par-
ent and daughter elements is possible. For
wider use, it was necessary to develop tech-
niques for the accurate measurement of
these elements in common rocks, where the
concentration may be only several parts
per million. The method of isotopic dilu-
tion proved satisfactory and permitted ex-
tension of the technique to other radio-
active elements such as potassium and ru-
bidium ("Determination of thorium by
isotope dilution"). The increased sensitiv-
ity of the isotopic dilution method makes
possible for the first time age determina-
tions by several radioactive decay systems
on a single rock, thus providing an oppor-
tunity to compare the reliability of the
various decay systems for dating the rock.
When the most reliable methods are as-
certained, a large variety of geochemical
and geophysical problems can be studied.
In addition, the availability of source-dated
minerals permits a study of natural nuclear
physical rate processes such as branching
ratios, neutron flux in rocks, and natural
fission.

The program, which has been executed

GEOPHYSICAL LABORATORY

in part, consists first in collecting large
quantities of a few common rocks from
which the constituent minerals suitable for
age determination may be easily isolated
("Preparation of specimens"). The min-
erals which have been separated include
zircon, sphene, apatite, biotite, lepidolite,
and perthite. A known weight of each
mineral, to which have been added known
amounts of stable isotopes, is then decom-
posed by appropriate reagents. When the
desired elements have been separated from
the resulting homogeneous solution by spe-
cial processes, they are analyzed by mass
spectrometer, and each element's absolute
concentration in the mineral is thereby de-
termined. The program includes the de-
termination of U 238 , Pb 206 , Pb 207 , Pb 208 ,
Th 232 , Rb 87 , Sr 87 , K 40 , Ca 40 , and A 40 for
each mineral. The ages based on the emis-
sion of alpha particles are given by the
ratios Pb 207 /Pb 206 ("Lead-lead ages"),
Th 232 /Pb 208 ("Thorium-lead ages"), and
U 238 /Pb 206 ("Uranium-lead ages"). Those
ages based on beta particle emissions are
given by the ratios Rb 87 /Sr 87 ("Rubidium-
strontium ages"), K 40 /Ca 40 , and K 40 /A 40 .
Considerable progress has been made in
the age-determination program in spite of
the many technical difficulties. The ages
based on alpha emissions have been ob-
tained for several minerals in three gran-
ites. As yet the ages for the same minerals
based on beta emissions have not been done,
but the techniques are now being worked
out on more satisfactory material for which
the age is also of considerable interest. The
results obtained cast some doubt on the
assumption that the age of a rock can be
determined solely from a single mineral
("Transfer of lead in the Essonville gran-
ite"). In addition, the ambiguity of the re-
sults from the various lead age techniques
points up the necessity for further work
on the source of errors, but final analysis

of the many problems encountered must
await the results of the entire program.

Determination of thorium by isotope di-
lution. The technique which has made the
present age studies possible is described in
detail for the element thorium. The stable
isotopes necessary for the practical execu-
tion of the technique have been made avail-
able by the Atomic Energy Commission.

Thorium in thorium-bearing minerals
has but one isotope, as shown in figure
13a. In order to determine thorium by
the method of isotope dilution, another
isotope of thorium in milligram quanti-
ties is needed. Th 230 is a daughter of U 238 ,
of sufficiently long half life to be present in
nearly pure form in originally thorium-free
uranium minerals. Since Th 230 is not yet
available from the AEC, about 1 mg of
such thorium was extracted from an old
pitchblende and made into the carrier solu-
tion. The isotopic composition of the car-
rier material is shown in figure 13^; the
ratio Th 230 /Th 232 is about 4. The carrier
solution is calibrated from this ratio and
from the ratio of a mixture of a known
volume of the carrier solution to a known
mass of Th 232 . The isotope distribution of
the mixture is shown in figure i3<r. The
thorium in a rock is determined by equi-
librating a known volume of carrier solu-
tion with a solution containing all the
thorium of the specimen. A fraction of
the thorium sufficient for mass spectro-
metric analysis is then extracted and ana-
lyzed, and the isotope distribution so ob-
tained (fig. i^d) is used to calculate the
thorium content of the rock or mineral.
The method is applicable to all elements
which have two stable or long-lived iso-
topes, and is only slightly more complex
when the element to be analyzed has
several isotopes.

Preparation of specimens. An extensive
field trip to the mapped Pre-Cambrian

CARNEGIE INSTITUTION OF WASHINGTON

areas of southern, southwestern, and west-
ern United States was made by a party of
Institution geologists, chemists, and physi-
cists to collect granites and pegmatitic min-
erals for laboratory investigation. Granites
were collected from sixteen significant lo-
calities, and lepidolites were obtained from

and their support and assistance proved to
be invaluable in obtaining representative
materials.

Thin sections of all rock samples were
prepared for petrographic examination,
and the granites were crushed for mineral
separation. Combinations of magnetic and

6 r

M
ID

Z
- 3

I-
Z
u
(T

S3 i

230 232
NORMAL THORIUM

230

232

Th 230 CARRIER

230 232

0.100 ml CARRIER
SOLUTION

0.0946 9 CAPE ZIRCON

Fig. 13. Mass spectra for determination of thorium by the method of isotope dilution

232
3.496/ NORMAL Th

+
0.100 ml CARRIER
SOLUTION

five pegmatites, four of which were seem-
ingly associated with granite bodies. Ac-
companying minerals such as feldspar, bio-
tite, muscovite, amblygonite, beryl, and
tourmaline were also collected at the peg-
matite localities for laboratory evaluation
as to their suitability either for age de-
termination or for measurement of isotope
ratios. Wherever possible, geologists work-
ing in the particular areas were consulted,

density separations were employed to yield
pure samples of the constituent minerals.
In some unfavorable cases 50 pounds of
rock yielded but 100 mg of zircon, barely
enough for preliminary analysis. Mineral
fractions of zircon, sphene, and biotite
were extracted from a granite collected at
Cape Town, Union of South Africa, and
from a heavy mineral concentrate from a
granite in Southern Rhodesia, obtained

GEOPHYSICAL LABORATORY gl

through the co-operation of Dr. A. M. nearly intact owing to its younger age and
MacGregor, of the Rhodesian Geological lower radioelement concentration. Since
Survey. zircon is one of the most stable minerals

Lead-lead ages. The lead analyses are known, the Cape zircon therefore repre-
the collaborative work of Dr. Claire C. sents one of the most favorable subjects for
Patterson, of the California Institute of thorium-lead age measurement on a rock
Technology. The leads of three granites mineral. It is seen that the thorium-lead
have been studied during the past year: and uranium-lead ages agree much more
Essonville, Ontario; Uncompahgre, Colo- closely for the Cape than for the Essonville
rado; and Cape Town, Union of South zircon. Allowing for the fact that the
Africa. The ages obtained from these Cape lead data are still of a preliminary
measurements are given in column 2 of nature, this agreement indicates that tho-
table 9. The values have an analytical ac-
curacy of about 5 per cent, unless other-
wise indicated. The Cape zircon ages are
based on preliminary data and are accurate
to about ±10 per cent.

Thorium-lead ages. The striking fea-
ture of the thorium-lead ages is the fact
that they are all much lower than the
uranium-lead and lead-lead ages of the
same minerals (table 9) . This discrepancy
has been found by other workers for a
number of the uranium-thorium ores. In
particular, five of the six monazite age de-
terminations reported in the literature re-
semble the data of table 9 in having low
thorium-lead ages as compared with the

uranium-lead ages, but no adequate expla- rium age discrepancies in zircons may be
nation has been offered for these observa- a function of the interrelated quantities,
tions. Since it now appears that many the radioelement concentration, and the
granitic minerals can be expected to yield extent of crystal damage,
low thorium-lead ages, it is probable that Another interesting result which may
the goal of future thorium work will be contribute to our understanding of this
more to determine the cause of the dis- problem has been obtained by leaching the
crepancy than to measure ages by this Essonville sphene with cold 6N hydro-
method. Some progress has been made chloric acid and analyzing the solution,
with the problem during the past year, al- The solution was found to contain 20 per
though the data are too few to permit cent of the sample's uranium, 40 per cent
generalization. of its thorium, and 70 per cent of its lead.

The comparison between the Essonville Moreover, the acid removed 65 per cent
and Cape zircons, however, is illuminat- of the radiogenic Pb 208 , but only 25 per
ing. X-ray examinations showed that the cent of the radiogenic Pb 206 . Thus for this
crystal structure of the Essonville zircon is mineral the lead generated by thorium is
badly shattered by radiation damage (met- more susceptible to removal than that gen-
amict), whereas that of the Cape zircon is erated by uranium. These observations

TABLE 9

Lead

AGES OF

GRANITIC MINERALS

Age (million years)

Sample

U 238 /Pb ;

206 Pb2"VPb 206 Th

232 /Pb 208

Essonville

granite:

Zircon ....

1030

1060

415

Sphene. . . .

920=b80 1060±200

202

Uncompahgre

granite:

Apatite. . . .

1050

1810±160

200

Biotite

3200

1700

1200

Cape granite:

Zircon ....

347

550

278

CARNEGIE INSTITUTION OF WASHINGTON

might indicate that loss of Pb 208 rather
than addition of thorium is the cause of
the low thorium-lead age for the mineral.
It is possible to lose Pb 208 without a pro-
portionate loss of Pb 206 , if there is an in-
homogeneous distribution of uranium and
thorium in the crystals.

In the case of the Essonville zircon it has
been found that a hot aqua regia leach will
reduce the alpha activity of the mineral by
35 per cent. This leach solution will be
studied in the same manner as the sphene
leach solution, in an effort to obtain further
information on the problem of low tho-
rium-lead ages.

Uranium-lead ages. The uranium-lead
age of the Cape zircon (table 9) is less
than the lead-lead age by an amount that
is far outside the limits of analytical er-
ror. It is believed that the lead-lead age is
the more nearly correct and that the ura-
nium-lead age is low. The Essonville
zircon occurred in the rock as large crys-
tals several millimeters in diameter,
whereas Cape zircon occurred as small
crystals but a few hundreths to a tenth of
a millimeter in diameter. The hypothesis
now favored is that the smaller size of
the Cape zircon crystals permitted lead
loss while the larger size of the Esson-
ville crystals prevented it. We have no
data for the diffusion rate of lead in zir-
con, but data exist for the diffusion of
lead in a perthite feldspar which indicate
that it could diffuse in the mineral over
distances of approximately 0.1 mm in one
billion years at 300° C under the influence
of a concentration gradient.

Rubidium-strontium ages. The rubid-
ium-strontium measurements have been
extended to include some 25 lithium micas,
obtained from pegmatites located in Pre-
Cambrian areas of South Africa and North
America. Table 10 gives the measurements
on South African lepidolites, and table 11

those for North America. The absolute
ages are determined from a decay con-
stant for Rb 87 of 1.13X io -11 yr _1 , the aver-
age of two recent determinations by ab-
solute beta counting. There is a possibility
that this decay constant may be revised in
the future if the existence of an unusual

TABLE 10

Apparent ages of South African
lepidolites

Age
Pegmatite location (million

years)

Popes Claim, S. Rhodesia 374°

Letaba, Transvaal 3850

Hombolo, Tanganyika 3 2 5°

M'bale (1), Uganda 2460

M'bale (2), Uganda 2370

Lunya, Uganda 2 34°

Kinderzitt, Namaqualand 1210

Jakkalswater, Namaqualand 1100

Muika (1), Belgian Congo 1220

Muika (2), Belgian Congo 1210

Karibib-Usakos II, SW. Africa 600

Sahatany, Madagascar 690

Alto Ligonha, Mozambique 570

Dogon Daji, Nigeria 625

TABLE 11

Apparent ages of North American
lepidolites

Age
Pegmatite location (million

years)

Bagdad, Arizona 2660 ±260

Bonneville, Wyoming 3570 ± 300

Dixon, New Mexico 3450 ±400

Ohio City, Colorado 3440 ± 300

type of radioactivity — "bound beta decay"
— can be proved. Diffusion of rubidium or
radiogenic strontium out of the crystal lat-
tice would also lead to erroneous ages. Loss
of rubidium would increase the measured
age.

The values presented in table 10 are simi-
lar to some of those published in last year's

GEOPHYSICAL LABORATORY

report. The ages of specimens collected
from different pegmatites in a geologically
homogeneous area agree within 10 per cent.
In one instance, however, the disagreement
between these measurements and published
syntheses of African Pre-Cambrian tecton-
ics is so striking that the need for caution
in interpreting rubidium-strontium ages is
again highlighted.

The results in table 11 suggest the exist-
ence of a Pre-Cambrian area in the United
States as ancient as any in North America.
Geographically, the three pegmatites with
ages of about 3500 million years are located
roughly between the 2000-million-year-old
pegmatites of the Black Hills, reported last
year, and the 2700-million-year-old pegma-
tite at Bagdad, Arizona. If other dating
methods support these interpretations, the
work will have important bearing on theo-
ries of continent formation. The existence
of very ancient minerals in the United
States as far south as New Mexico is impor-
tant to theories of continent formation.
These measurements are also a part of the
study of the correlation of ages determined
by different radioactive decay systems.

An attempt was made this year to see
whether the large variation in the isotopic
abundances of the isotopes of strontium
thought to be stable could be due to radio-
active decay of zirconium or uranium
contained in the mineral zircon. The ab-
solute amount of strontium in the blank
run in parallel with the zircon was 2.5 ±
0.5 micrograms. That found in one gram
of zircon was within 20 per cent of this
amount. This can be considered good
agreement between two blank runs. It is
therefore concluded that the radioactive
decay of zirconium or uranium in the
amounts present in the mineral biotite
could not contribute in any way to the
variation of the ratio Sr 86 /Sr 88 found in
biotite. It is of interest to note that the
strontium determination in zircon was

made using Sr 87 from lepidolite as carrier
material. This isotope's purity in the stron-
tium in lepidolite is so high (99 per cent)
that it is ideal for the determination of
small amounts of common strontium. It is
expected that rubidium-strontium measure-
ments will be made in the near future on
the three granites for which lead ages were
determined.

Transfer of lead in the Essonville gran-
ite. Further work on the Essonville granite
has given evidence of a transfer of lead
(and possibly uranium and thorium)
within the minerals of the rock. The

T'ABLE 12
Isotopic composition of essonville leads

Sample 206/204 207/204 208/204

Perthite from granite

(observed) 18.4 15.6 39

Granite composite (cal-
culated) 16.4 15.3 32

Perthite from pegma-
tite (observed) 16.7 15.0 35

perthite present in the rock contains so
much lead as compared with uranium and
thorium that the isotopic composition of
the lead would not have been affected by
radioactive decay within the mineral dur-
ing the billion years the granite has ex-
isted. Thus it should be the lead present
when the rock was formed. When one
compares this measured isotopic composi-
tion with that calculated from the known
age and from uranium, thorium, and lead
data for the rock, it is found that the cal-
culated ratios of Pb 206 , Pb 207 , and Pb 208 to
Pb 204 are substantially lower (see table 12).
This indicates either addition of radio-
genic lead to the feldspar or else a frac-
tionation of lead from uranium and tho-
rium in the rock as a whole.

In view of this information, a large
perthite crystal from a pegmatite in the
Essonville area was analyzed for the iso-

CARNEGIE INSTITUTION OF WASHINGTON

208

topic composition of its lead. Diffusion with high ratios of Pb 206 , Pb 207 , and Pb

processes should have had much less effect to Pb 204 ) from the sphene of the granite

here, since the crystal was several centime- with cold 6N hydrochloric acid. Such a

ters in length. The pegmatite lead agrees process could serve as the source of the lead

very well with that calculated from the which has been added to the perthite.

total granite. It is accordingly believed If such transfers of atoms between the

that there has been no measurable frac- various minerals of granites prove to be

tionation of uranium, thorium, and lead the general case, the interpretation of age

in the rock as a whole and that there has measurements based on these minerals will

been addition of radiogenic lead to the be greatly complicated. The studies of the

feldspar since it was formed. It is possible coming year should give an indication of

to leach highly radiogenic lead (i.e., lead how important this factor will be.

MISCELLANEOUS INVESTIGATIONS

The system Na 2 — CaO — Si0 2 (Morey).
Years ago, a fundamental paper on this
system was published by Morey and
Bowen (1925, No. 580), and later addi-
tional results were published by Morey
(1930, No. 720). In connection with a pa-
per on the optical properties and densities
of soda — lime — silica glasses, written in col-
laboration with Merwin (1932, No. 802),
a large number of additional mixtures
were prepared and liquidus determinations
were made on these mixtures. These re-
sults, which have not been published, do
not greatly alter the positions of the bound-
ary curves. A doubt was cast, however,
on the existence of the compound Na 2 0*
2CaO*3Si0 2 , and it is desirable to pub-
lish the additional proof which has been
obtained. More experiments should be car-
ried out in the high-soda part of the sys-
tem. Some time ago, evidence of the exist-
ence of a compound of the composition
Na 2 0*CaO was found, and additional
work has been done by hydrothermal
methods on its occurrence.

New techniques for accurate measure-
ment of lattice parameters on the powder
spectrometer (Smith). Two advances in
practical details have been made. First, it
has been found that high-accuracy meas-
urements can be taken from simple smear

mounts by using thin smears containing a
calibrating material such as quartz or sili-
con. Secondly, it has been possible consid-
erably to reduce the amount of material
required for the pressed mounts by minor
variations of the technique described by
Adams and Rowe (paper in press).

A systematic method of determining the
lattice parameters from the observed dif-
fraction angles has been devised in collabo-
ration with Chayes. The angles are con-
verted into (2-values by the method of
Donnay and Donnay. Qk — ^h%kai, where
hue are integers appropriate to the ^th re-
flection and cti are the required lattice pa-
rameters. If K values of Qk are measured,
then a set of K simultaneous equations can
be set up from which the six ais can be de-
termined. The least-square solution is ob-
tained by setting up the matrix \hik\,
computing the corresponding inverse ma-
trix, and then using this to compute the
lattice parameters. The inverse matrix was
kindly calculated for us by the Applied
Mathematics Division of the National Bu-
reau of Standards. The two advantages of
the method are: (1) the most accurate so-
lution is obtained in a direct systematic
way, and (2) the same inverse matrix may
be used for different crystals if they give
the same measurable X-ray reflections.

GEOPHYSICAL LABORATORY

This condition holds for several mineral
series, such as the plagioclases, to which
the method is being applied at the mo-
ment. The use of the same inverse matrix
eliminates a great deal of tedious compu-
tation and should result in encouraging
investigators to make more accurate meas-
urements of more crystals.

Field studies of metamorphic roc\s
(Yoder). In the summer of 1952 the clas-
sic metamorphosed areas in the Grampian
highlands of Scotland, the Oslo Fjord re-
gion in Norway, the Falun region in Swe-
den, and the Orijarvi region in Finland
were visited in the company of their most

distinguished investigators. The major
purposes of these general surveys were:
(1) to learn the results of the newer de-
tailed studies in these and other compara-
ble regions, (2) to discuss the concept of
the "water-deficient region" in terms of
the field observations, (3) to collect series
of rocks in well established zones of pro-
gressive metamorphism, and (4) to evalu-
ate the possibility of carrying out the
principal metamorphic reactions experi-
mentally. All purposes were achieved. The
results will be incorporated in the near
future with discussions of the program to
which they apply.

The "Summary of Published Work" be-
low briefly describes the papers published
in scientific journals during the report
year. In addition, the following papers are
now prepared for publication: L. H.
Adams and F. A. Rowe, "The prepara-
tion of specimens for the focusing-type
X-ray spectrometer"; L. H. Adams, M. A.
Tuve, and H. E. Tatel, "Studies of the
earth's crust using waves from explosions";
F. Chayes, "A test of the revised deter-
minative chart for plagioclase"; F. Chayes,
"In defense of the second decimal"; F.
Chayes, "The theory of thin section analy-
sis"; G. W. Morey, "Hydrothermal syn-
thesis"; G. W. Morey, "Silica and sili-

cates"; G. W. Morey, "The binary system
K4P2O7— KPOs"; G. W. Morey, "The bi-
nary system NaPOs— KPO3"; G. W.
Morey, "The system H 2 0— NaPOs"; J. F.
Schairer, "The system K2O — MgO — Si0 2 :
I. Results of quenching experiments on
four joins in the tetrahedron cordierite —
forsterite — leucite — silica, and on the join
cordierite — mullite — potash feldspar"; J. V.
Smith, "Re-examination of the crystal
structure of melilite"; O. F. Tuttle and
M. L. Keith, "Quartz and feldspar of a
Tertiary granite in relation to its thermal
history"; J. Van den Heurk, "Improved
hydrothermal quenching apparatus."

SUMMARY OF PUBLISHED WORK

(1167) Change of melting point of diopside with
pressure. H. S. Yoder, Jr. Jour. Geol.,
vol. 60, pp. 364-374 (1952).

The melting point of synthetic diopside
under pressures up to 5000 bars was found
by experiment to be well represented by the
equation t m = 1391.5 + 0.01297P, where t m is
the melting point in degrees centigrade and
P is the pressure in bars. The character of
the melting curve supports the prevailing
view that the melting curve rises indefinitely
with depth in the earth.

The considerable volume change involved
in melting is emphasized in a quantitative re-
view of the process of magma production and
extrusion.

(1168) The atomic arrangements and bonds of

the gold-silver ditellurides. G. Tunell

and L. Pauling. Acta crystallogr., vol. 5,

PP- 375-38i (1952).

The structures of sylvanite, calaverite, and

krennerite, the three gold-silver ditellurides

occurring naturally, are basically similar, al-

CARNEGIE INSTITUTION OF WASHINGTON

though they have three different space-group
symmetries. In each of the three minerals the
gold and silver atoms are surrounded by six
tellurium atoms, which, however, are not all
at the same distance. In sylvanite and cala-
verite, each tellurium atom is surrounded by
three gold or silver atoms and three tellurium
atoms. In krennerite, part of the tellurium
atoms are surrounded by three gold or silver
atoms and three tellurium atoms, part by one
gold or silver atom and five tellurium atoms,
part by five gold or silver atoms and one tel-
lurium atom. An analysis is given of the
interatomic distances in terms of the resonat-
ing-valence-bond theory.

(1169) A test of the accuracy of chemical analy-
sis of silicate rocks. H. W. Fairbairn
and J. F. Schairer. Amer. Mineralogist,
vol. 37, pp. 744-757 (1952).

A six-component silicate glass of the pro-
portions found in granite was prepared and
its composition controlled so that significant
errors were eliminated except for Si0 2 (0.02
per cent) and Al 2 O s (0.01 per cent). Analy-
sis of this glass by eleven chemical labora-
tories gave mean values which showed satis-
factory agreement (<o.i per cent) for MgO,
CaO, Na 2 0, K 2 0. For Si0 2 and Al 2 O s a
reciprocal systematic error appeared whereby
Si0 2 was 0.4 per cent low and A1 2 3 was 0.6
per cent high; total Si0 2 and A1 2 3 agreed
almost exactly (<o.i per cent) with the glass
standard. The practical problem of evaluat-
ing error in a single analysis is discussed
briefly in its relation to these results.

(1170) The system sodium metaphosphate — cal-
cium metaphosphate. G. W. Morey. Jour.
Amer. Chem. Soc, vol. 74, p. 5783
(1952).

A study of the binary system sodium meta-
phosphate — calcium metaphosphate showed
the formation of a compound 2Na 2 0*CaO-
3P 2 5 with a congruent melting point at
738 ° C. The eutectic between Na 2 0-P 2 5
and 2Na 2 ■ CaO ■ 3P 2 5 is at 625 ° and about
1 per cent CaO-P 2 5 ; that between 2Na 2 0*
CaO-3P 2 5 and CaOP 2 5 is at 725 ° and
46 per cent CaO-P 2 5 .

(1171) The 10 per cent CaAl 2 Si 2 O g plane in the
system CaSiO s — Ca 2 Al 2 Si0 7 — NaAlSi0 4
— CaAl 2 Si 2 O s . H. S. Yoder, Jr. Jour.
Geol., vol. 60, pp. 586-593 (1952).

The high-temperature equilibrium relations
in the 10 per cent CaAl 2 Si 2 8 plane in the
system CaSi0 3 — Ca 2 Al 2 Si0 7 — NaAlSi0 4 —
CaAl 2 Si 2 8 have been investigated by the
quenching method. The results contribute to
the understanding of the origin and differen-
tiation of alkaline rocks.

(1172) The flow of glass at room temperatures.
G. W. Morey. Jour. Optical Soc. Amer.,
vol. 42, pp. 856-857 (1952).

A 60-inch plane mirror at the Mount Wil-
son Observatory, which weighed about 2000
pounds and was 9 inches thick, stood on edge
for 25 years. A careful study of the surface
gave no evidence of a distortion's having oc-
curred as large as a quarter of a wave length
of light. This affords a convincing demon-
stration that glass does not flow at ordinary
temperatures under its own weight. The
stresses produced by poor annealing, however,
are probably much greater than those pro-
duced by the disk, and the consensus of ex-
perience, empirical though it may be, is that
glass will flow at room temperatures under
stresses of the magnitude produced by poor
annealing.

(1173) The use of ion exchange columns in
mineral analysis for age determination.
L. T. Aldrich, J. B. Doak, and G. L.
Davis. Amer. Jour. Sci., vol. 251, pp.
377-387 (1953).

Ion-exchange columns of synthetic organic
resins have been found extremely useful in
the routine separation of the alkali elements
from those of the alkaline earths in the pro-
portions in which they are found in potas-
sium minerals. Complete separation of these
groups of elements is essential for the mass
spectrometric analysis of calcium and stron-
tium in rubidium-containing potassium min-
erals. Partial separations of rubidium from
potassium and strontium from calcium
which increase the relative concentration of
rubidium and strontium 100-fold are made

GEOPHYSICAL LABORATORY

8 7

without difficulty. The use of suitable com-
plexing agents for iron and aluminum per-
mits a high yield of calcium and strontium to
be obtained from very small samples of iron
aluminum silicate minerals.

(1174) List of systems investigated at Geophysi-
cal Laboratory. L. H. Adams. Amer.
Jour. Sci., Bowen Volume (vol. 250-A),
pp. 1-26 (1952).

The systems investigated at the Geophysi-
cal Laboratory during the more than forty
years of its existence have been arranged and
tabulated, together with the pertinent journal
references. It is believed that this list, con-
taining over 300 entries, will be found con-
venient by students of experimental petrology.

(1175) The differentiation of a composite aplite
from the Pribilof Islands, Alaska. T. F.
W. Barth. Amer. Jour. Sci., Bowen Vol-
ume (vol. 250-A), pp. 27-36 (1952).

The mode of occurrence and the field re-
lations of an aplite intrusion are illustrated.
The chilled margins show a high content of
alkalies and silica, but a lower content of
aluminum, iron, magnesium, and calcium,
than does the coarse-grained, granular central
portion. This arrangement may be explained
on the theory of thermodiffusion as described
by W. Wahl, whereas crystallization differ-
entiations would have resulted in the opposite
arrangement.

(1176) Chemical petrology of some metamor-
phosed Adirondack gabbroic, syenitic and
quartz syenitic rocks. A. F. Buddington.
Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 37-84 (1952).

The chemical petrology of 16 metamor-
phosed Adirondack gabbroic rocks is dis-
cussed in the light of 15 new chemical anal-
yses of their minerals, comprising 9 of horn-
blende, 5 of garnet, and 1 of augite. Four of
the rock analyses are new. There is a direct
sensitive correlation between the composition
of the rocks and the composition of their
minerals, consistent with an origin through
mineral reconstitution with little change in
bulk chemical composition. The nature and

composition of the minerals also depend on
whether the rock was reconstituted under
relatively dry conditions or in the presence
of hydrous fluids carrying fluorine or chlorine
or both.

Chemical aspects of the metamorphism of
some charnockitic syenitic and quartz syenitic
rocks are also discussed in the light of new
chemical analyses of 6 specimens of the major
rocks, 4 related small pegmatite seams, and
2 garnets. The pegmatite seams are in large
part garnetiferous and the product of local
metamorphic recrystallization with some met-
amorphic differentiation.

The metamorphism of the gabbroic, sy-
enitic, and quartz syenitic rocks occurred in
connection with regional high-grade dynamo-
thermal metamorphism, in large part in the
presence of intergranular films of fluid re-
lated to a subjacent rising granitic magma.
The granite magma later intruded the meta-
morphosed rocks and locally developed a su-
perimposed reconstitution on the early facies.
The rocks were in part metamorphosed under
conditions of the upper temperature range
for the amphibolite facies and in part in the
granulite facies. A geologic map showing the
regional variation in the nature and intensity
of metamorphism is given. The problem of
the garnets at the Barton Mine is discussed
and reasons are given for preferring an origin
as an aspect of regional metamorphism to an
origin through local recrystallization directly
consequent on syenite intrusion, as advocated
by some other students.

(1177) Relations between composition and in-
dices of refraction in natural plagioclase.
F. Chayes. Amer. Jour. Sci., Bowen Vol-
ume (vol. 250-A), pp. 85-105 (1952).

New determinative charts for the refractive
index-composition relation are presented. Pos-
sible discontinuities in the relation are dis-
cussed. Though complete optical continuity
is not disproved by the data, sample descrip-
tions based on one or two breaks in the series
suggest simple relations between the com-
monly measured natural plagioclases and the
albite and anorthite polymorphs produced by
laboratory synthesis.

CARNEGIE INSTITUTION OF WASHINGTON

(1178) Two new crystalline phases of the anor-
thite composition, CaO • A1 2 3 • 2Si0 2 .
G. L. Davis and O. F. Tuttle. Amer.
Jour. Sci., Bowen Volume (vol. 250-A),
pp. 107-114 (i95 2 )-

Two new crystalline phases of the composi-
tion CaAl 2 Si 2 8 are described, giving the
optical properties and X-ray powder diffrac-
tion data. Their synthesis and stability are
discussed. Comparisons are drawn with simi-
lar modifications of BaAl 2 Si 2 O s .

(1179) The symmetry change in the high-tem-
perature alkali-feldspar series. G. Don-
nay and J. D. H. Donnay. Amer. Jour.
Sci., Bowen Volume (vol. 250-A), pp.
1 15-132 (1952).

Accurate unit-cell dimensions are given for
n synthetic alkali feldspars ranging from
monoclinic KAlSi 3 8 to triclinic NaAlSi 3 O s .
Cell edges and angles are plotted against
composition. The only discontinuity is found
in the first derivative of the interaxial angle a
at the point where the change of symmetry
takes place. No two-phase region was ob-
served. The phase transition is a high-order
one, and, for practical purposes, solid solu-
tion may be considered complete.

(1180) On the granulites of Lapland. P. Eskola.
Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 133-171 (1952).

The granulites of Lapland are garnet-bear-
ing acid quartz-feldspar rocks, and mostly hy-
persthene-bearing intermediate, basic, and ul-
trabasic rocks, the latter forming sharply
bounded bands in the former. Twenty rock
analyses and 13 mineral analyses are given,
and a granulite norm is proposed to show the
critical minerals of the granulite facies. Spe-
cific properties of the granulite minerals are
described. A greater part of the granulites is
chemically aluminous rock, a fact which sug-
gests original argillaceous character. Graphite-
granulites show a composition like that of
sapropelic sediments; their carbon is supposed
to be of organic origin. The basic granulites
are interpreted as metamorphic volcanic or
hypabyssic rocks, whereas more coarse-grained

and homogeneous quartz-dioritic rocks are
assumed to be primary magmatic. The granu-
lite facies appear to be identical with that fa-
cies represented by the charnockites, the spe-
cific characteristics of the foliated granulites
being due to their tectonic history.

(1181) Orthopyroxenes of the Bushveld type,
ion substitutions and changes in unit cell
dimensions. H. H. Hess. Amer. Jour.
Sci., Bowen Volume (vol. 250-A), pp.
173-187 (1952).

The a, b, and c dimensions of the unit cell
of a number of orthopyroxenes of the Bush-
veld type were measured. The variation in
size with substitution of Fe +2 for Mg +2 and
Al +3 for Mg +2 and Si +4 are given. The optical
properties N e> 2V, and birefringence have
been revised and density determinations made
on a number of samples. For ordinary ortho-
pyroxenes of the Bushveld type with 2%
per cent R 2 3 , unit cell dimensions, N z , bire-
fringence, and density vary as linear func-
tions within the limits of accuracy of the ob-
servations, and thus appear to obey Vegard's
law. The quantities present of all of the main
constituents of these pyroxenes, magnesium,
iron, aluminum, and calcium, can be deter-
mined from optical property and unit cell
measurements without chemical analysis. It
is suggested that, other things being equal,
the aluminum content may increase with
pressure and should be further investigated
as a potential means of assessing depth of
crystallization.

(1182) Twinning frequency in feldspar pheno-
crysts from a quartz latite sill at Sierra
Blanca, Texas. E. Ingerson. Amer. Jour.
Sci., Bowen Volume (vol. 250-A), pp.
189-202 (1952).

Study of some 1500 sanidine phenocrysts
from a quartz latite sill in Texan Moun-
tain, Trans-Pecos, Texas, shows the following
abundances of (1) untwinned crystals, (2)
Carlsbad, (3) Manebach, and (4) Baveno
twins, respectively: from a zone near the up-
per contact, (1) 84.35 per cent, (2) 8.5 per
cent, (3) 6.9 per cent, (4) 0.25 per cent; from

GEOPHYSICAL LABORATORY 89

the center of the sill, (1) 94.3 per cent, (2) version studies on synthetic quartz. For ex-

3.3 per cent, (3) 2.4 per cent, (4) not ob- ample, synthetic quartz grown in the pres-

served. The average size of the phenocrysts ence of germanium has an inversion break

near the top of the sill is larger than that of 40 ° C above the normal range for quartz,

those near the center; the twinned crystals whereas quartz grown in the presence of

are significantly larger than the untwinned lithium and aluminum has the inversion tem-

ones in both places. These facts are tenta- perature lowered by as much as 120 ° C. Spec-

tively explained as due to (1) somewhat trochemical analysis of natural quartz with

earlier initiation of crystallization of twinned different inversion temperatures shows varia-

crystals, (2) floating of early-formed pheno- tions in the amounts of minor elements, and

crysts, and (3) greater tendency for nuclei to this variation is in turn accompanied by

develop twinning in the more disturbed con- changes in the cell dimensions as measured by

tact zones. Spectrographic and chemical anal- X-ray methods.

yses show significantly more rubidium and The amount of solid solution is influenced

lithium in the untwinned and more of the al- by the temperature of growth, and therefore

bite molecule in the twinned crystals. Both the inversion temperature can be used as an

these relations would be expected if the indication of the relative temperature of for-

twinned crystals started forming earlier and mation of samples of quartz which grew in

at a higher temperature than the untwinned similar chemical environments. A general

ones. relation between the inversion temperatures

and the occurrence of natural quartz is de-

(1183) Significance of variation in the high-low rived from comparison of occurrences for

inversion of quartz. M. L. Keith and which approximate formation temperatures

O. F. Tuttle. Amer. Jour. Sci, Bowen can De inferred. For example, quartz from

Volume (vol. 250-A), pp. 203-280 (1952). rhyoliteS) where the presence o£ g l ass indi .

The high-low quartz inversion of some 250 cates high temperature, inverts at lower tem-

quartz specimens has been studied by a dif- peratures than quartz from veins or from

ferential thermal method. The method does cavities in limestone, where geologic evidence

not permit determining the exact temperature indicates a lower temperature of formation,

at which the inversion would take place un- There are, however, numerous exceptions

der equilibrium conditions, but the measured to this relationship, and the temperature dif-

temperature is probably not greatly different ferences in the inversion can generally be

from the equilibrium value. The difference taken to indicate only that the quartz is in

between the inversion temperatures of various some way different. In other words, the in-

quartz specimens can, however, be deter- version can in most cases be used only as a

mined to within ±o.i° C when the heat ab- "fingerprint" method of comparing quartz

sorption is rapid (approximately 90 per cent from various sources and for the study of zon-

of all specimens studied). ing within rock bodies; in these applications

Differences between the inversion tempera- it has proved to be potentially useful. A large

tures of natural quartz samples vary over a proportion of the present report is a prelimi-

range of 38 ° C, and the inversion of synthetic nary "fingerprinting" of quartz from various

quartz varies over a range of 160 C. Over rock types.
95 per cent of all natural specimens examined,

however, invert on heating within a range (1184) Japanese twins of quartz. S. Kozu.

of 2.5 ° C. Amer. Jour. Sci., Bowen Volume (vol.

Variations found in the inversion tempera- 2 5°-A), pp. 281-292 (1952).

ture are believed to be the result of solid solu- The possible types of quartz twins which

tion of small amounts of ions other than Si 4+ incorporate the "Japanese twin" law are enu-

and O 2- . This conjecture is supported by in- merated. Etching was used to determine the

9 o

CARNEGIE INSTITUTION OF WASHINGTON

twin law and composition plane. Differences
in the crystals are related to their genetic
conditions.

(1185) Thermochemistry of plagioclase and al-
kali feldspars. F. C. Kracek and K. J.
Neuvonen. Amer. Jour. Sci., Bowen Vol-
ume (vol. 250-A), pp. 293-318 (1952).

Employing a hydrofluoric acid solution cal-
orimeter, previously described, measurements
of the heats of solution (decomposition) were
performed in 20 per cent hydrofluoric acid at
74.7 ° C for 21 feldspar materials in the
plagioclase and alkali feldspar series. In com-
bination with other pertinent measurements,
these will lead to evaluation of the heats of
formation of the feldspars. The results pre-
sented in this paper show that plagioclase
feldspars in their natural state exist not as
the simple, continuous solid-solution series
that is found at the liquidus, but rather fall
into three groups with dividing lines drawn
at 30 and 65-70 mole per cent anorthite. The
intermediate labradorite compositions appear
to be mixtures of feldspars of the limiting
compositions indicated. The alkali feldspars
have thermal properties highly dependent on
their origin and thermal history. Values of
the latent heats of melting of albite and
anorthite have been deduced from the ther-
mochemical measurements. The average val-
ues are 49.2 cal/g for albite and 64.7 cal/g
for anorthite. The heat of inversion of low,
natural albite to high-temperature albite is
13 cal/g.

(1186) The effect of temperature on the sym-
metry of high-temperature soda-rich feld-
spars. W. S. MacKenzie. Amer. Jour.
Sci., Bowen Volume (vol. 250-A), pp.
319-342 (1952).

X-ray and optical data show that natural
anorthoclase and synthetic soda-rich feldspars,
triclinic at room temperature, acquire mono-
clinic symmetry on heating. On cooling, the
reverse process takes place. The experimental
results suggest that there are two triclinic
forms of synthetic high-albite and adjacent
feldspars. The temperature at which the sym-
metry becomes monoclinic differs in the two

forms, and there are slight differences in the
X-ray powder diffraction patterns at room
temperature.

The inversion temperatures of natural an-
orthoclases may prove useful in geologic ther-
mometry.

(1187) The system H 2 0— Na 2 0— Si0 2 at 400 ° C.
G. W. Morey and J. M. Hesselgesser.
Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 343-371 (1952).

A method and an apparatus have been de-
vised which have made possible the analysis
of coexisting gas and liquid phases in the
system H 2 — Na 2 — Si0 2 at 400 ° C and at
pressures up to 2500 bars. The isothermal
polybaric saturation curves — both gaseous and
liquid — of quartz, sodium disilicate, and so-
dium metasilicate have been determined, and
the melting pressure of Na 2 • 2Si0 2 has been
found to be 2200 bars. The boundary curve
G + L '+ quartz + sodium disilicate is con-
tinuous and does not intersect the critical end-
point curve, and probably the boundary curve
G + L + sodium disilicate + sodium meta-
silicate has a similar relationship. There are
probably two critical regions surrounded by
critical curves, one of which is represented by
(G = L) + Na 2 0-Si0 3 , the other by (G =
L) + Na 2 0*2Si0 2 , although the possibility
has not been excluded that these two critical
curves intersect, giving rise to a double criti-
cal end point (G = L) + Na 2 OSi0 2 +
Na 2 # 2Si0 2 . There is also a critical region
represented by (G = L) + Qtz.

(1188) Heat of formation of merwinite and
monticellite. K. J. Neuvonen. Amer.
Jour. Sci., Bowen Volume (vol. 250-A),
PP. 373-38o (1952).

The heats of formation of merwinite,
Ca 3 Mg(Si0 4 ) 2 , and monticellite, CaMgSi0 4 ,
have been measured by means of solution
calorimetry. Natural merwinite from Crest-
more, California, and natural monticellite
from Magnet Cove, Arkansas, were used for
the measurements. The results, corrected for
theoretical compositions of the two minerals,
are for merwinite AH = —57,020 ±380 cal/
mole and for monticellite AH = — 27,560 ±

GEOPHYSICAL LABORATORY

140 cal/mole. The values account for the
heat of formation of the two silicates at 25 ° C
from quartz, calcium oxide, and magnesium
oxide (prepared by dehydrating magnesium
hydroxide 2.5 hours at 1425 C).

(1189) The chemistry of the Keweenawan lavas.
P. Niggli. Amer. Jour. Sci., Bowen Vol-
ume (vol. 250-A), pp. 381-412 (1952).

Research by various authors and especially
by T. M. Broderick and Henry R. Cornwall
has done much to clarify the chemistry of the
Keweenawan lavas. Calculations and graphic
representations based on molecular units in-
stead of weight percentages are carried out in
the present paper with a view to clarifying
the differentiation tendencies in these basic
magmas. It is pointed out that the formation
of the so-called pegmatite layers can be re-
garded as the first indication of a differentia-
tion tending toward the development of a
spilitic to keratophyric association.

(1190) The system diopside — forsterite — anor-
thite. E. F. Osborn and D. B. Tait.
Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 413-433 (1952).

The system diopside — forsterite — anorthite
is an example of a system containing a pyrox-
ene, an olivine, and a feldspar in which vir-
tually no solid solution or reaction relation
among the phases exists. This system, how-
ever, may be considered to be one face of a
tetrahedron representing the system diopside
— forsterite — anorthite — silica, which is note-
worthy for crystal-liquid reaction phenomena
as liquids crystallize under equilibrium condi-
tions. Phase-equilibrium data are presented,
and their bearing on the crystallization of
spinel from basaltic liquids and on the forma-
tion of coronite is discussed.

(1191) A reconnaissance of liquidus relations in
the system K 2 0*2Si0 2 — FeO — Si0 2 . E.
Roedder. Amer. Jour. Sci., Bowen Vol-
ume (vol. 250-A), pp. 435-456 (1952).

A reconnaissance has been made of that
portion of the system K 2 — FeO — Si0 2 hav-
ing K 2 0:Si0 2 ratios greater than 1:2. The
results on 62 compositions are presented as a
preliminary phase diagram for the condensed

subsystem K 2 • 2Si0 2 — FeO — Si0 2 , repre-
senting liquidus relations in equilibrium with
metallic iron under one atmosphere pressure
of pure nitrogen. Two new compounds,
K 2 0-FeO-3Si0 2 and K 2 • FeO • 5Si0 2 , were
found, and their relations to equivalent com-
pounds in the systems K 2 — A1 2 ; , — Si0 2
and K 2 — MgO — Si0 2 discussed.

(1192) Leucite, potash nepheline, and clino-
pyroxene from volcanic lavas from south-
western Uganda and adjoining Belgian
Congo. Th. G. Sahama. Amer. Jour.
Sci., Bowen Volume (vol. 250-A), pp.
457-470 (1952).

Four specimens of rocks from the young
volcanic belt of southwestern Uganda and
adjoining Belgian Congo have been investi-
gated. Leucite, nepheline, and clinopyroxene
(diopside), extracted from the rocks by means
of heavy liquids, have been analyzed chemi-
cally. Optical properties are also given.
Chemical compositions of the analyzed min-
erals are discussed.

(1193) The system FeO— A1 2 3 — Si0 2 . J. F.
Schairer and K. Yagi. Amer. Jour. Sci.,
Bowen Volume (vol. 250-A), pp. 471-
512 (1952).

Phase-equilibrium studies of this system, at
and just below temperatures where a liquid
phase is present, show one ternary compound,
the iron analogue of cordierite (2FeO*
2Al 2 3 , 5Si0 2 ), which decomposes at 1210
±io° C to mullite, tridymite, and liquid.
The fields of stability of corundum, mullite,
hercynite, iron cordierite, cristobalite, tridy-
mite, fayalite, and wiistite have been deline-
ated. The following invariant points (the
first two eutectics, and the remainder reaction
points) were located: fayalite + wiistite -f-
hercynite + liquid, ii48°±5°; fayalite +
iron cordierite + tridymite + liquid, 1083
it 5 ; fayalite + iron cordierite + hercynite
+ liquid, io88°±5°; hercynite + iron cor-
dierite + mullite + liquid, i205°±io°;
iron cordierite + mullite + tridymite +
liquid, 1210 ±io°; corundum + mullite
+ hercynite + liquid, i38o°±5°; cristo-
balite + tridymite + mullite + liquid,

o . o

1470 ± 10 .

9 2

CARNEGIE INSTITUTION OF WASHINGTON

Iron cordierite crystallizes with some re-
luctance, and the metastable invariant points
fayalite + tridymite + spinel + liquid at
io73°±5° and mullite + hercynite + tridy-
mite + liquid at 1205 ° ±io° can be realized.
No ferrosilite or almandine garnet could be
crystallized from the melts at any tempera-
ture, even when melts were seeded with these
crystalline phases. Natural almandine from
Botallack, England (91.3 per cent almandine)
when heated begins to decompose at an ap-
preciable rate as low as 900 ° C, and yields a
mixture of hercynite, iron cordierite, and fa-
yalite. The bearing of these results on petrol-
ogy and slags is discussed.

(1194) Single-crystal measurements on paracel-
sian. J. V. Smith. Amer. Jour. Sci.,
Bowen Volume (vol. 250-A), pp. 513-
515 (1952).

The cell dimensions and space group of
paracelsian have been determined, and have
been compared with those of topaz and dan-
burite. There is a close similarity between
paracelsian and danburite.

(1195) Temperature scales and silicate research.
R. B. Sosman. Amer. Jour. Sci., Bowen
Volume (vol. 250-A), pp. 517-528 (1952).

The temperature scale for the range 300 ° C
to the melting point of platinum, used by the
Geophysical Laboratory from 19 12 to the
present, was based on the nitrogen thermome-
ter work of Day and Sosman. The first
International Temperature Scale (1927) was
a thermodynamic scale, with fixed points a
little higher than those of Day and Sosman.
The revision of the I. T. S. adopted in 1948,
by raising the value of the constant in the
Wien-Planck law of radiation, brings the
high-temperature scale nearer to the Geo-
physical Scale. Curves show the differences
between the Geophysical and the I. T. S. of
1927 and 1948.

(1196) Some trends of basaltic magma in lime-
stone syntexis. C. E. Tilley. Amer. Jour.
Sci., Bowen Volume (vol. 250-A), pp.
529-545 (1952).

The trend of basaltic magma consequent
on limestone reaction is considered in the

light of new data from the contact zones of
Camas M6r, Muck, and Scawt Hill, County
Antrim, in the British Tertiary igneous prov-
ince. The chain of reaction products — meli-
litic theralites, melilitic nepheline dolerites,
and the derivative nepheline dolerites — ex-
presses the trend of a residual liquid in which
the concentration of iron relative to magnesia
is strongly emphasized. In this and other
features these assemblages present a contrast
to the more common mafic alkali rocks of
the type of nepheline basalt, nephelinite, and
their melilitic and monticellite-bearing varie-
ties. Some reference is made to the genetic
features of these latter rocks and to the prob-
lem of the desilication of granitic liquid by
limestone syntexis.

(1197) The angle between the #-axis and the
trace of the rhombic section on the
(oio)-pinacoid in the plagioclases. G.
Tunell. Amer. Jour. Sci., Bowen Volume
(vol. 250-A), pp. 547-551 (1952).

The angle between the a-axis and the
trace of the rhombic section on the {010}-
pinacoid is given in terms of y, the angle
between the tf-axis and the £-axis, and the
angle between the {ooi}-pinacoid and the
(oio)-pinacoid by the formula

cos (001 A 010)

cot o =

cot y

Values of a for a number of plagioclases from
pure albite to pure anorthite were calculated
from the interaxial angles of Wiilfing by
means of this formula. The values of o for
these plagioclases calculated from the inter-
axial angles of Wiilfing by Rosenbusch and
Miigge were computed by means of an incor-
rect formula.

(1198) Optical studies on alkali feldspars. O. F.
Tuttle. Amer. Jour. Sci., Bowen Volume
(vol. 250-A), pp. 553-567 (1952).

Optical properties of alkali feldspar pheno-
crysts from extrusive rocks place these feld-
spars, sanidines, and anorthoclases in a series
with high-temperature albite as the soda end
member. This information, together with the
results of Spencer (1937), permits classifying
the alkali feldspars into four series.

GEOPHYSICAL LABORATORY

(1199) The MgO— A1 2 3 — Si0 2 — H 2 system
and the related metamorphic facies.
H. S. Yoder, Jr. Amer. Jour. Sci., Bowen
Volume (vol. 250-A), pp. 569-627 (1952).

Equilibria in a portion of the system MgO
— ALO3 — Si0 2 — H 2 have been determined
at temperatures from 430 ° to 990 ° C at pres-
sure of water vapor up to 30,000 pounds per
square inch. The univariant pressure-temper-
ature curves have been bracketed for the fol-
lowing reactions: brucite + clinochlore ^
forsterite + spinel + vapor, talc + clino-
chlore <=± forsterite + cordierite + vapor,
clinochlore ^± forsterite + cordierite + spi-
nel + vapor; and some of the previously in-
vestigated reactions in the MgO — SiOo — H 2
system are confirmed.

A new low-temperature form having the
clinochlore composition was prepared, but its
stability has not been fixed with certainty.
It is suspected that this form may be isostruc-
tural with one form of serpentine. Pyrope
was not synthesized, and natural pyrope was
not found to be stable in the presence of an
excess of water vapor in the pressure and
temperature range investigated. Amesite and
magnesian chamosite were not produced.
Metastable growth of many of the phases
occurring in the system is common, and they
survive long periods of time.

The phase assemblages obtained in equi-
librium with vapor are summarized at sig-
nificant temperature intervals at 15,000 psi.
These assemblages outline compositions in the
system for which water is prohibited as a
stable phase. The conclusion is reached that
the presence of an "excess" or "deficiency"
of water vapor greatly influences the mineral-
ogy of a metamorphic rock. On this basis,
it is demonstrated that all the now accepted
critical assemblages which define the meta-
morphic facies may have formed under the
same pressure and temperature conditions,
the different facies being primarily a func-
tion of the bulk composition. A new set of
critical assemblages indicative of significant
pressure and temperature conditions is re-
quired for the various stages in progressive
metamorphism. In addition, it is shown that

a common hydrothermal alteration which has
been interpreted as retrograde metamorphism
need not mean a change in pressure or tem-
perature conditions, but may mean an access
of water vapor.

(1200) Bowen Volume, American Journal of
Science, vol. 250-A. viii + 627 pp. (1952).

Bound volume comprising papers nos. 1174
through 1 199.

(1201) A note on the stability of jadeite. L. H.
Adams. Amer. Jour. Sci., vol. 251, pp.
299-308 (1953).

By use of certain simplifying assumptions,
the stability pressure, at various temperatures,
for the formation of jadeite from nepheline
and albite is calculated from thermal data.
The pressure increases by about 18 bars per
degree; and at 400 ° C the pressure is esti-
mated to be 4800 bars. From the known be-
havior of hydrates under high pressure it is
suggested that under pressure analcite may
melt incongruently to form jadeite and water,
so that at a sufficiently high pressure jadeite
rather than analcite would be produced from
a mixture of the correct composition, even
in the presence of excess water.

(1202) Polymorphism. F. C. Kracek. Encyclo-
paedia Britannica, vol. 18, pp. 187^-188
(1953 copyright).

A review of polymorphism in its classic
aspects.

(1203) The relation between area and volume in
micrometric analysis. F. Chayes. Min-
eralogical Mag., vol. 30, pp. 147-149
0953).

The validity of a modal analysis as an
estimate of the composition by volume of the
parent rock has nothing to do with the fact
that it is also, and quite inadvertently, an
estimate of the composition by volume of the
thin section on which it is made. The only
role of the thin section is to provide a surface
on which the areas of the phases can be meas-
ured, and in this respect, its only advantage
over an opaque polished surface arises from

CARNEGIE INSTITUTION OF WASHINGTON

the ease and accuracy with which the phases
can be identified by transmitted light.

The basic relation between area and vol-
ume in micrometric analysis seems to have
been clearly understood by Delesse. In intro-
ducing the method more than a century ago,
he stated very firmly that no attempt should
be made to "see into" the rock; measure-
ments were to be confined to surface areas.
His demonstration is based largely on geo-
metric intuition, and the intuition of succeed-
ing generations of petrographers has not been
so clear as his. The lack of adequate analyti-
cal proof has been responsible for much sub-
sequent confusion.

(1204) Research in experimental geology at the
Geophysical Laboratory. J. F. Schairer.
Capital Chemist, vol. 3, pp. 130-133
(i953).
This paper, prepared for the May 1953 is-
sue of the Capital Chemist, which featured
the Geophysical Laboratory, broadly sketches
the Laboratory's history and major accom-
plishments. Emphasis is placed on the quan-
titative approach through experimental ap-
plication of chemistry, physics, physical chem-
istry, and thermodynamics to geologic prob-
lems concerning the nature and formation of
igneous and metamorphic rocks. Before such
studies could be properly undertaken, the
necessary apparatus and methods had to be
developed. Among these preliminary efforts
were the establishment of a standard tempera-
ture scale for temperatures at least up to
1755 ° C and the development of the method
of quenching for the accurate determination
of phase-equilibrium relations. Since silica is
one of the major constituents of the earth's
crust, much of the Laboratory work has in-
volved study of polycomponent silicate sys-
tems, both anhydrous and hydrous. From
these studies, our knowledge of the melting
and crystallization relations of the various
minerals has been established and extended.
In addition, valuable data immediately ap-
plicable to industrial purposes have been ob-
tained. For example, contributions have been
made directly to the Pordand cement, glass,

ceramic, steel, and metallurgical industries.
The chemist and physicist have gone into the
field with the geologist to learn his problems
and then have set about to find some of the
answers through an experimental approach.

(1205) Annual report of the Director for 195 1-
1952.

(1206) Petrochemical studies on the alkalic rocks
of the Morotu district, Sakhalin. K. Yagi.
Bull. Geol. Soc. Amer., vol. 64, pp. 769-
809 (i953)-

The alkalic rocks of the Morotu district
occur as sheets, laccoliths, and dikes in the
Tertiary formations and are probably Plio-
cene. The rocks vary from dolerite, the most
abundant, through monzonite to syenite.
Most of the intrusive bodies, except small
sheets and dikes, are composed of dolerite in
the marginal parts and monzonite or syenite
or both in the central parts, with gradational
boundaries. Monzonite and syenite also occur
as irregular veinlets or schlieren in the doler-
itic parts, with sharp boundaries. Thus differ-
entiation in situ is strikingly displayed. These
intrusive bodies are, therefore, formed by
composite intrusion. Contact effects of the
intrusions on the Tertiary formations are
weak and are confined to the immediate con-
tact zone. Phenomena of assimilation have
not been observed.

Principal minerals of these rocks are plagio-
clase, anorthoclase, microperthite, olivine, ti-
tan- and alkali-pyroxenes, alkali-amphiboles,
titanbiotite, analcite, and iron ores. Feldspar
evolves from labradorite to oligoclase and is
joined by anorthoclase and microperthite in
the later stage. The trend is always toward
enrichment in the orthoclase molecule. Py-
roxene, the most important mafic mineral,
forms a continuous reaction series from diop-
sidic augite through titanaugite, soda-augite,
and aegirinaugite to aegirine. Amphibole also
forms a reaction series from barkevikite
through kaersutite and hastingsite to arfved-
sonite, but it is not clear whether their relation
is continuous or discontinuous. Biotite is
abundant in the dolerites, but very low in the
syenites. Analcite is abundant in all rocks,

GEOPHYSICAL LABORATORY

apparently formed hydrothermally in the later
stages.

Chemically these rocks range from 46 to
61 per cent Si0 2 . Their alkali-lime index is
50.0, placing the suite in the "alkalic series."
The parental magma is inferred to have been
olivine basaltic magma with moderately al-
kalic affinity. The various rock types are
believed to have been formed chiefly by crys-
tallization differentiation from the parental
magma after it intruded into the present posi-
tion. Volatile components have also played
an important role during crystallization, espe-
cially in the later stage.

There is no absolute enrichment in iron
throughout the crystallization course of the
Morotu rocks. It is inferred that not only the
shape of the intrusive bodies, but also the
composition of the parental magma deter-
mines whether iron enrichment or alkali and
silica enrichment will prevail in the intrusive
bodies.

Its geographical position, its period of ac-
tivity, and its petrological character justify the
conclusion that the Morotu district constitutes
the northeastern end of the Circum-Japan Sea
province of Cenozoic alkalic rocks.

(1207) The crystal structure of paracelsian,
BaAl 2 Si 2 O g . J. V. Smith. Acta crystal-
logr., vol. 6, pp 613-620 (1953).

Paracelsian, BaAl 2 Si 2 8 , is isostructural with
danburite, CaB 2 Si 2 8 , whose structure is sim-
ilar to that of the feldspars. It is monoclinic
but very nearly orthorhombic. The struc-
ture was determined by two-dimensional
Fourier methods. Fairly accurate dimensions
of the barium polyhedra were obtained. The
value for the mean distance Sii /2 Ali /2 -0 is
of interest. Correlation with recent values for
Si-O, Si 3 / 4 Al 1 /4-0 suggests values of i.6o 5 A
for Si-O and 1.78 + 0.02 A for Al-O. The
latter value agrees well with the sum of the
Goldschmidt radii but is larger than the sum
of the Pauling radii.

BIBLIOGRAPHY

Adams, L. H. List of systems investigated at
Geophysical Laboratory. Amer. Jour. Sci.,
Bowen Volume (vol. 250-A), pp. 1-26
(1952).

A note on the stability of jadeite. Amer.

Jour. Sci., vol. 251, pp. 299-308 (1953).
Aldrich, L. T., J. B. Doak, and G. L. Davis.
The use of ion exchange columns in mineral
analysis for age determination. Amer. Jour.
Sci., vol. 251, pp. 377-387 (i953).
Barth, T. F. W. The differentiation of a com-
posite aplite from the Pribilof Islands,
Alaska. Amer. Jour. Sci., Bowen Volume
(vol. 250-A), pp. 27-36 (1952).
Buddington, A. F. Chemical petrology of some
metamorphosed Adirondack gabbroic, sye-
nitic and quartz syenitic rocks. Amer. Jour.
Sci., Bowen Volume (vol. 250-A), pp. 37-84
(1952).
Chayes, F. Relations between composition and
indices of refraction in natural plagioclase.
Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 85-105 (1952).

The relation between area and volume

in micrometric analysis. Mineralogical Mag.,
vol. 30, pp. 147-149 (i953).
Davis, G. L., and O. F. Tuttle. The new crys-

talline phases of the anorthite composition,
CaO • Al 2 O s • 2Si0 2 . Amer. Jour. Sci., Bowen
Volume (vol. 250-A), pp. 1 07-1 14 (1952).
— See Aldrich, L. T.

Doak, J. B. See Aldrich, L. T.

Donnay, G., and J. D. H. Donnay. The sym-
metry change in the high-temperature alkali-
feldspar series. Amer. Jour. Sci., Bowen Vol-
ume (vol. 250-A), pp. 1 15-132 (1952).

Donnay, J. D. H. See Donnay, G.

Eskola, P. On the granulites of Lapland. Amer.
Jour. Sci., Bowen Volume (vol. 250-A), pp.
133-171 (1952).

Fairbairn, H. W., and J. F. Schairer. A test
of the accuracy of chemical analysis of sili-
cate rocks. Amer. Mineralogist, vol. 37, pp.

744-757 (1952).

Hess, H. H. Orthopyroxenes of the Bushveld
type, ion substitutions and changes in unit
cell dimensions. Amer. Jour. Sci., Bowen
Volume (vol. 250-A), pp. 173-187 (1952).

Hesselgesser, J. M. See Morey, G. W.

Ingerson, E. Twinning frequency in feldspar
phenocrysts from a quartz latite sill at
Sierra Blanca, Texas. Amer. Jour. Sci.,
Bowen Volume (vol. 250-A), pp. 189-202
(1952).

9 6

CARNEGIE INSTITUTION OF WASHINGTON

Keith, M. L., and O. F. Tuttle. Significance of
variation in the high-low inversion of quartz.
Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 203-280 (1952).

Kozu, S. Japanese twins of quartz. Amer. Jour.
Sci., Bowen Volume (vol. 250-A), pp. 281-
292 (1952).

Kracek, F. C. Polymorphism. Encyclopaedia
Britannica, vol. 18, pp. 1870-188 (1953 copy-
right).

and K. J. Neuvonen. Thermochemistry

of plagioclase and alkali feldspars. Amer.
Jour. Sci., Bowen Volume (vol. 250-A), pp.
293-318 (1952).

MacKenzie, W. S. The effect of temperature on
the symmetry of high-temperature soda-rich
feldspars. Amer. Jour. Sci., Bowen Volume
(vol. 250-A), pp. 319-342 (1952).

Morey, G. W. The flow of glass at room tem-
peratures. Jour. Optical Soc. Amer., vol. 42,
pp. 856-857 (1952).

The system sodium metaphosphate — cal-
cium metaphosphate. Jour. Amer. Chem.
Soc, vol. 74, p. 5783 (1952).

and J. M. Hesselgesser. The system

H 2 0— Na 2 0— Si0 2 at 400 ° C. Amer. Jour.

Sci., Bowen Volume (vol. 250-A), pp. 343-

371 (1952).
Neuvonen, K. J. Heat of formation of mer-

winite and monticellite. Amer. Jour. Sci.,

Bowen Volume (vol. 250-A), pp. 373-380

(1952).

See Kracek, F. C.

Niggli, P. The chemistry of the Keweenawan

lavas. Amer. Jour. Sci., Bowen Volume (vol.

250-A), pp. 381-412 (1952).
Osborn, E. F., and D. B. Tait. The system

diopside — forsterite — anorthite. Amer. Jour.

Sci., Bowen Volume (vol. 250-A), pp. 413-

433 (1952).

Pauling, L. See Tunell, G.

Roedder, E. A reconnaissance of liquidus rela-
tions in the system K 2 • 2Si0 2 — FeO — Si0 2 .
Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 435-456 (1952).

Sahama, Th. G. Leucite, potash nepheline, and
clinopyroxene from volcanic lavas from
southwestern Uganda and adjoining Belgian
Congo. Amer. Jour. Sci., Bowen Volume
(vol. 250-A), pp. 457-470 (1952).

Schairer, J. F. Research in experimental ge-
ology at the Geophysical Laboratory. Capi-
tal Chemist, vol. 3, pp. 130-133 (1953).

and K. Yagi. The system FeO — A1 2 3 —

Si0 2 . Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 471-512 (1952).
See Fairbairn, H. W.

Smith, J. V. Single-crystal measurements on
paracelsian. Amer. Jour. Sci., Bowen Vol-
ume (vol. 250-A), pp. 513-515 (1952).

The crystal structure of paracelsian,

BaAl 2 Si 2 O g . Acta crystallogr., vol. 6, pp.
613-620 (1953).

Sosman, R. B. Temperature scales and silicate
research. Amer. Jour. Sci., Bowen Volume
(vol. 250-A), pp. 517-528 (1952).

Tait, D. B. See Osborn, E. F.

Tilley, C. E. Some trends of basaltic magma
in limestone syntexis. Amer. Jour. Sci.,
Bowen Volume (vol. 250-A), pp. 529-545
(1952).

Tunell, G. The angle between the a-axis and
the trace of the rhombic section on the
(oio)-pinacoid in the plagioclases. Amer.
Jour. Sci., Bowen Volume (vol. 250-A), pp.

547-551 (1952).

and L. Pauling. The atomic arrange-
ments and bonds of the gold-silver ditel-
lurides. Acta crystallogr., vol. 5, pp. 375-
381 (1952).

Tuttle, O. F. Optical studies on alkali feldspars.
Amer. Jour. Sci., Bowen Volume (vol.
250-A), pp. 553-567 (1952).

See Davis, G. L.; Keith, M. L.

Yagi, K. Petrochemical studies on the alkalic
rocks of the Morotu district, Sakhalin. Bull.
Geol. Soc. Amer., vol. 64, pp. 769-809

(i953).

See Schairer, J. F.

Yoder, H. S., Jr. Change of melting point of
diopside with pressure. Jour. Geol., vol. 60,
pp. 364-374 (1952).

The 10 per cent CaAl 2 Si 2 8 plane in

the system CaSiO s — Ca 2 Al 2 Si0 7 — NaAlSi0 4
— CaAl 2 Si 2 8 . Jour. Geol., vol. 60, pp. 586-

593 (1952).

The MgO— A1 2 3 — Si0 2 — H 2 system

and the related metamorphic facies. Amer.
Jour. Sci., Bowen Volume (vol. 250-A), pp.
569-627 (1952).

DEPARTMENT OF TERRESTRIAL MAGNETISM

Washington, D. C.
MERLE A. TUVE, Director

Several items of particular interest and our nuclear physicists have elected to de-
satisfaction are included in the following vote themselves to the understanding and
report, which is in effect a running account experimental clarification of biological
of one year of work by the physicists who problems is only an illustration of the char-
comprise the stafT of this Department, acter of basic research. Strict lines of de-
Their work is not directed toward the ex- marcation no longer exist between subjects
ploration of a selected set of particular of study when the mode of approach is that
mysteries connected with the earth's mag- of fundamental science. Physics is a mother
netism, although problems in this area are subject for a whole range of specialized
not ignored. inquiries. It is encouraging to find in this

The work of the investigators at the De- age of great specialization that many of

partment today is the present expression of the keenest minds in basic research in vari-

a policy on basic research evolved through ous laboratories no longer accept the tra-

several decades. The original plans of the ditional restriction of their research interest

Department for a magnetic survey of the and activity to a single narrow set of ques-

earth led in a few years to the formulation tions, but are viewing their own special

of some basic questions about the structure advances in the perspective of a whole

of the earth and its atmosphere, and the range of related and active research areas,

various observed effects of the sun. By the Research, as a professional activity, is

middle twenties the conviction grew that vastly different in this mid-century decade

a healthy program on basic questions in from the minor adjunct of university and

geophysics could best be fostered by having hospital programs which it was in the

a group of laboratory physicists concerned United States when the Carnegie Institu-

with fundamental problems, such as nu- tion and this Department were organized

clear physics, working in close daily as- shortly after 1900. The great proliferation

sociation with the men concerned with the of research and development laboratories

geophysical questions. in industrial establishments, colleges, uni-

About half the staff members are ac- versities, and agencies of government

tively directing their efforts toward fur- which took place during and just after

ther illumination of matters relating to the World War II has brought into unmis-

physical and geological history of the earth takably prominent relief the question

or to the state of the upper atmosphere, or whether direct contribution to a selected

to solar and cosmic influences observed on technical task or subject, such as terrestrial

the earth. These investigators are chiefly magnetism, is in fact the most fruitful

men whose training and part of whose type of contribution the Institution may

efforts are concerned with modern labora- attempt to foster, a question to which our

tory physics, and their daily companions, experience in previous decades had given

at lunch and in seminar discussions, are at least a partially negative answer. The

men who are intensively working on lab- staff of the Department undertakes in-

oratory problems. The fact that several of stead to make a continuous evaluation of

9 8

CARNEGIE INSTITUTION OF WASHINGTON

the most creative and fruitful activities
they can individually carry on as physi-
cists devoted to fundamental research,
moving forward of course from their own
special fields of training and interest, and
hence with a natural bias toward geophysi-
cal problems. This evaluation and guid-
ance of emphasis is not made in any de-
tached or distant fashion, but by actually
living and thinking and working together,
individually dedicating their efforts to
problems in their own fields of compe-
tence which seem both important and
approachable, but without undue preoc-
cupation with some preselected and highly
specific topic.

The continuing value of such an evolv-
ing and experimental example of a small
and traditional research laboratory, as con-
trasted with the expansive and highly goal-
directed activities which receive such em-
phasis in the physics departments of our
universities today, is clear to anyone who
quietly contemplates the present confusion
of physics with technology and the wide-
spread public misconceptions which em-
phasize utility as the aim of basic research.
The latter error appears in some measure
to be connected with the budget defense
of large-scale activities.

Research here is regarded as a highly
personal activity, not a large-scale effort
for a team. There are important values
for each investigator, however, in the stim-
ulus of close association with a few other
intensely active and interested research
men, several of them usually working on
problems closely akin to his own, others
equally active and enthusiastic about prob-
lems in a widely different area, so that in
the course of each week or two he encoun-
ters the fresh stimulus of new ideas or re-
search progress in a whole range of scien-
tific fields. This is a satisfying way for a re-
search man to live, and it is fruitful. A
concrete object lesson to the staff members

and to many visitors has been the striking
contrast between the actual operations here
and current notions about "team research"
and the supposed necessity for a research
man to be extremely specialized and nar-
row in all his interests and thinking.

The nuclear physics men who have con-
cerned themselves with measuring radio-
active isotope relationships by means of
the mass spectrograph have found stron-
tium-rubidium ratios in the minerals sepa-
rated from rocks collected in New Mexico,
Colorado, and Wyoming which show, if
isotope ratios are reliable measures of age,
that these rocks were laid down about 3400
million years ago. Certain parts of Canada,
Africa, and Scandinavia have long been
known as areas of the most ancient igne-
ous activity, and their minerals show ages
of the same magnitude. It will be a sur-
prise to many that rocks of such extreme
apparent age are found among the granites
and pegmatites of the Rocky Mountains.

There has been an unexpected develop-
ment of new evidence concerning the im-
portance of atmospheric circulation and
zonal winds in the production of magnetic
storms. The further elaboration of the dy-
namo theory, made possible by modern
studies of winds in the upper atmosphere,
coupled with recent revisions in our under-
standing of the electrical conductivity of
the ionosphere overhead, now indicates
that the electric currents which produce
magnetic storm effects are primarily in
the atmosphere, and the wind motions
which produce them are initiated by radia-
tions and perhaps by particles from the
sun.

Other items of growing future interest
concern developments in radio astronomy,
and measurements on the pathways in
living things by which inorganic carbon
is incorporated to form particular groups
of amino acids and amino acid residues
in proteins and nucleic acids.

DEPARTMENT OF TERRESTRIAL MAGNETISM

It may well be imagined that the dis- basic research in the physical sciences to-

cussions of the physicists who are sharing day. It is hoped that this report, in addi-

ideas and hopes and successes and disap- tion to presenting factual material, may

pointments as they work together here also serve as some indication of the

illustrate in a modest but personal way the variety and the friendly intensity of those

stimulus and satisfaction connected with discussions.

EXPERIMENTAL GEOPHYSICS

RADIO ASTRONOMY

R. J. Britten, L. Owren, F. G. Smith, H. E.
Tatel, M. A. Tuve, and H. W. Wells

For nearly thirty years the Department
has been actively engaged in studies of the
upper atmosphere using radio waves. The
original approach, which began in the au-
tumn of 1924, laid the foundation for the
enormous proliferation of pulse radio, with
its applications to radar and communica-
tion, in addition to providing highly analyt-
ical procedures for the study and predic-
tion of radio communication pathways in
their relation to solar activity. The estab-
lishment of the Central Radio Propaga-
tion Laboratory at the National Bureau of
Standards for systematic ionosphere studies
over our part of the globe relieved the De-
partment in 1946 of its far-flung responsi-
bilities for ionosphere operations. The de-
cision was made to emphasize research
activities, and hence small-scale operations,
since research is invariably a personal ac-
tivity of an individual or of two or three
individuals with closely related fields of
interest but different immediate problems.

Radio "noise" signals which appeared to
come from the plane of the galaxy were
observed by Jansky, of the Bell Telephone
Laboratories, in the early 1930's, and these
observations were extended by Grote
Reber in Illinois roughly ten years later.
Radar interference from the sun during
the war led to an active program in
England and Australia for the study of
radio stars and of radio noise from active
spots on the sun, especially after the war.

Activities along these lines in the United
States have been somewhat less explora-
tory and of narrower scope. It has become
apparent, however, that the observational
material of radio astronomy has direct
bearing on other current interests within
the Institution, and relates directly to our
long series of studies on the upper atmos-
sphere. About two years ago Mount Wil-
son Observatory asked the Department, by
reason of its long experience with similar
electronic techniques, to develop a basis
for assessment of the observational mate-
rials and opportunities in radio astronomy
in relation to the more traditional astro-
nomical programs. A matter of particular
interest for the Department has been the
demonstrated relation of the scintillation
of radio stars to the turbulence and mo-
tions of ionized regions in the upper
atmosphere.

A vigorous start has been made during
the past eighteen months toward several
different types of specific studies in radio
astronomy as a new direction in the evolu-
tion of our upper-atmosphere studies.

Radio astronomy owes its existence as a
field of scientific endeavor to the fact that
radio waves of wave lengths between 1 cm
and 15 m which originate in celestial ob-
jects can penetrate the earth's atmosphere
and be detected from ground installations.
Research on radio emissions at centimeter
and decimeter wave lengths calls for differ-
ent techniques and engineering methods
from those required for research on emis-
sions at meter wave lengths. For the re-

100

CARNEGIE INSTITUTION OF WASHINGTON

ception of centimeter waves, radio tele-
scopes are generally used, that is, antennas
which are analogous to the reflecting tele-
scopes used in the visual and photographic
range of the spectrum. The large parabolic
antenna (plate i) used for study of the
hydrogen-line (21-cm) emission is an ex-
ample of the radio telescope. At the longer
wave lengths which are normally used for
studies of radio waves from the sun or
stars, however, a radio telescope having
resolution comparable to that of the instru-
ments used for centimeter waves would
take on rather formidable proportions. To
achieve the desired resolution, resort is had
to a radio analogue of the interferometer
technique used by Michelson and Pease
at the Mount Wilson Observatory for in-
creasing the resolving power of the 100-
inch telescope. Michelson's optical inter-
ferometer consists of two mirrors on a
long beam across the aperture of the 100-
inch reflector. The telescope is pointed
toward a giant star whose diameter is to
be measured, and the separation of the
two mirrors is increased until the inter-
ference fringes appearing at the eyepiece
disappear. In the radio interferometer, an-
tennas are substituted for Michelson's mir-
rors, and electronic devices replace the
100-inch telescope. A radio interferometer
permits determination of the intensity, the
location, and the angular diameter of a
radio source. Such sources may be discrete
galactic objects, extragalactic nebulae, the
sun, or active regions on the sun. Accurate
determination of the position of a radio
source with an interferometer requires an-
tennas oriented in different directions, usu-
ally east-west and north-south, with sepa-
rations of ten to several hundred wave
lengths. The position measurements re-
quire a precise knowledge of the position
of the interference lobes as well as a high
degree of stability in their direction. The
lobe positions are determined from reg-

ular calibrations, and the stability is
achieved through careful instrumentation
which eliminates or minimizes phase shifts
arising in the amplifiers or interconnecting
cables.

Hydrogen-Line Emission from
the Galaxy

In September 1951, announcement was
made of the experimental verification by
Purcell and Ewen at Harvard of line emis-
sion from atomic hydrogen at 1420.405
megacycles. This faint emission from hy-
drogen atoms, distributed with very low
density between the stars, had been pre-
dicted in 1944 by Van de Hulst in Hol-
land. The significant feature of this emis-
sion, which has been under observation in
Holland and Australia almost since its
first detection at Harvard, is that it is radi-
ated at a fixed atomic frequency, and
hence, by the Doppler effect, it is an excel-
lent measure of the velocity of hydrogen
gas clouds with respect to the earth. A
great many interesting problems relating
to the structure of our galaxy can be ap-
proached with this new observational tool,
and in fact an interpretation of the rela-
tively limited observations to date has al-
ready been made by Oort and others in
Holland to show the probable existence of
spiral arms in our own galaxy.

Observations of this hydrogen line were
begun at our Department in November
1952, using a stationary parabola (Wurz-
burg radar dish, on loan from the Na-
tional Bureau of Standards). The com-
plexities of interpretation due to the rela-
tively brief time a given area of the sky
(about two degrees in diameter) is under
observation, and the necessity for sweep-
ing the frequency through the expected
Doppler range, made it clearly necessary
to mount the receiving parabola on an
equatorial axis. This was done during the

Department of Terrestrial Magnetism

Plate i

Plate 2

Department of Terrestrial Magnetism

^sstc*

Hohn

. _ ,-^yy

«« m ynnw *- ^* ***?£Z£&^"

;=»=*

1 1 J ;

1 1 " * )

| ' - j \ V~|

*^^^

** v

#i* *"*

-3

3
3

(/3

<i;

4—1

C/3

1-1

t/3

CTj

M-l

*-)

6 T3

1-7 CJ

«^n fcJD

>% as

Ih Ih

2 u

_§ a

i-l 12

— 1 rt

«5

to
<u

n 3

•- 1 -n

g 3

3
3

'■H 3h

DEPARTMENT OF TERRESTRIAL MAGNETISM

101

winter, and observations began again at
the end of April. The electronic proce-
dures are complex, as the signal energy
from the hydrogen in space is, for the
most intense spots, only a little more than
one per cent of the random noise in the
radio receiving equipment. Measurement
of these hydrogen signals therefore re-
quires time averages over many seconds.
The accumulation of reliable observations
on this fascinating but faint light from
the incredibly tenuous gas between the
stars of our galaxy is necessarily a slow
process. As the report year ends, all de-
sign features have been worked out and
tested for a multiple-channel apparatus de-
signed to observe all Doppler frequencies
continuously, giving separate and simul-
taneous averaging on about 50 different
frequency channels. Construction of this
equipment will take several months, and
meanwhile observations, anchored to the
Dutch points on the galactic equator, are
being made of gas clouds in several speci-
fied directions.

Active Radio Sources on the Sun

A radio interferometer operating at 207
mc/sec (1.45 m) for solar work has been
nearly completed at the Derwood Experi-
mental Laboratory in Maryland. It will be
used for location of active regions on the
sun with an accuracy of one minute of arc
(1/30 of a solar diameter) and for study
of their minute-to-minute fluctuations in
position. The solar radio interferometer
will have two pairs of antennas on an east-
west line, spaced 50 and 170 wave lengths
apart, respectively, and one north-south
pair spaced at 50 wave lengths. An experi-
mental version of the east-west spacing of
50 wave lengths was first operating in
November 1952 (see plate 2), and recently
regular daily observations of the sun with
this unit were started. The increased in-

tensity of solar radio noise appears to origi-
nate in relatively small disturbance centers
associated with sunspots or other visible
events on the solar disk. Some preliminary
results have demonstrated the feasibility
of identifying active radio sources with
visible structures on the sun. It is expected
that intensified collaboration in radio and
optical observations will yield fruitful re-
sults leading to a better understanding of
the dynamic and physical properties of
the sun and its atmosphere.

Radio Star Scintillations

The fluctuations of radio stars have been
associated with irregularities in the earth's
ionosphere by British and Australian inves-
tigators. These scintillations were origi-
nally thought to be actual variations in the
intensity of the signals from the radio
stars. Simultaneous measurements of radio
star scintillations at stations separated by
a few miles, however, gave results which
demonstrated the existence of irregular re-
fraction processes in our outer atmosphere.
There is reason to believe that these fluctu-
ations are caused by irregularities in the
upper part of the F region of the iono-
sphere which are not observable by ordi-
nary methods of ionospheric sounding.
Studies of these radio star scintillations
promise to provide a new tool for investiga-
tion of the electrical properties of our at-
mosphere and for the observation of
"winds" or traveling disturbances at
heights which are otherwise inaccessible.

The star scintillation experiment which
is just getting under way will provide the
first information from this continent. Pri-
mary emphasis will be placed on the de-
termination of diurnal and seasonal char-
acteristics of the fluctuations at a frequency
near 40 mc/sec. The results will eventu-
ally be compared with those from other
locations, such as Great Britain, for assess-
ment of more global properties.

102

CARNEGIE INSTITUTION OF WASHINGTON

Subsequently, the experiments may ex-
pand to the operation of tripartite stations
at spacings of a few miles for measure-
ment of velocity and direction of the
upper-atmospheric "winds."

THE UPPER ATMOSPHERE

H. W. Wells

Traveling Disturbances

The ionospheric program of 1952 added
significantly to our knowledge of traveling
disturbances in the E and F regions of the
ionosphere. Still unanswered, however, is
the fundamental question whether these
phenomena are actual transport of ions as
by winds, or the result of wave motions or
the sweeping past of ionizing beams caus-
ing transient irregularities in the distribu-
tion of ionization. A significant fact which
is emerging from our observations is the
evidence that the F-region disturbances are
moving predominantly from west to east
at velocities of from 100 to 200 meters per
second. (The velocity of sound in the same
region is approximately 500 m/sec, and
the rotational velocity of the earth at this
latitude is approximately 350 m/sec, also
west to east.) These results, however, are
confined to daylight hours, since the local
ionospheric irregularities, which require
identification and precise timing at each
station of our observing triangle, lose most
of their identity at night.

The size of the local ionospheric disturb-
ances may be estimated from their veloc-
ity and duration. Many of the larger
"clouds" retain their identity for 30 to 40
minutes. Probable dimensions are there-
fore between 200 and 400 km. Smaller dis-
turbances, having quasi periods of 10
minutes or less, are often observed super-
imposed on the larger ones. These may
have dimensions of 50 to 100 km or less.
From a meteorological point of view, the
smaller disturbances often have the appear-

ance of turbulence within the larger clouds.
The observing program for 1953 pro-
vides for improved instrumentation giving
better time resolution. The three observ-
ing stations will operate within a radius
of 20 miles from the Derwood Laboratory,
but a fourth station will be added as a
check point. Efforts are being made to
simplify the problem of data analysis by
the operation of stations at selected fixed
frequencies, rather than sweep frequencies.
Some preliminary results from two sta-
tions 15 miles apart give promise of signifi-
cant improvement in ability to identify
and time the traveling disturbances with
precision.

Pen-and-Ink Ionospheric Recorders

The development of fixed-frequency
ionospheric sounders, designed to produce
pen-and-ink records of precise height and
echo amplitude, was continued into the
stage of practical field tests. One instru-
ment was in intermittent operation at Der-
wood for several months. Another instru-
ment was operated simultaneously at the
Poolesville (Maryland) site. The field
tests revealed an inherent susceptibility to
noise or other interfering signals which
reduced the useful portions of records
below our minimum requirements of per-
formance. Accordingly, the radiofrequency
components have been adapted to photo-
graphic recording from cathode-ray indi-
cators, and the modified instruments are
being used in the program for investiga-
tion of traveling ionospheric disturbances.

Forward Scatter Experiment

An experiment for determining the ef-
fectiveness of forward scatter from the F
region in supporting long-distance propa-
gation of ultra-high-frequency signals has
been pursued only intermittently because
of pressure of other interests. The develop-

DEPARTMENT OF TERRESTRIAL MAGNETISM IO o

ment of a high-power pulse transmitter at been primarily the accomplishment of a

43.5 mc was curtailed by delays in obtain- survey, except in limited regions; we have

ing the required special license. Subse- been concerned with the development of

quent field tests under conditions simulat- analytical procedures for obtaining and

ing forward scatter from the E region evaluating information about the structure

produced marginal signals except during of the outer crust of the earth. Over 2200

periods of meteoric activity. At such times, records at different observing sites from

the signals were heard distinctly in Florida, roughly 270 explosions, each using 1200 to

but reports from Peru were negative. It is 3000 pounds or more of high explosive,

planned to resume the experiment in the have been made in the course of our

fall of 1953, using a simplified transmitting studies. About a third of these are in the

antenna of high gain (corner reflector eastern United States, the rest in New

type). Mexico, California, Minnesota, and Puget

Some properties of meteor trails were Sound,
studied, using an application of the radio These studies have made it quite clear
interferometer technique. The meteor that the simplified textbook picture (de-
trails were illuminated by a beamed con- duced some years ago from near-earth-
tinuous-wave transmitter operating at quake studies) of the outer crust of the
Cedar Rapids, Iowa, on a frequency near earth, just below the sedimentary rocks, as
50 mc. The ionization produced by the a succession of granitic and basaltic layers
rapidly moving head of the meteor simu- underlain by a mantle of ultrabasic rock is
lated the moving-point source as studied hardly in accord with the complexities ac-
by interferometer techniques in radio as- tually found in the field. Using individual
tronomy. By use of a phase-sensitive de- instruments and arrays of instruments
tector it was possible to measure a compo- spread over 1 km or more, at locations de-
nent of velocity across the interferometer termined to within 0.1 km, and with the
pattern. Similarly, the path length of the timing of the shot and the travel times
meteor in the ionized medium may be of the waves accurate to within 0.02 second,
measured by the number of interference we find real fluctuations in time and ampli-
fringes produced. Estimates of meteoric tude which must be ascribed to a variety of
velocities and path length were found to structures in the crust and not simply to
be consistent with results obtained from the variations in rock structure within a
other sources. few hundred feet of the instruments.

Vigorous efforts have failed to find any

THE EARTH'S CRUST indication of intermediate horizontal layers

Seismic Studies °^ basaltic rocks, and in fact earthquake

observations are clearly not suited to defin-

H. E. Tatel and M. A. Tuve ing such layers as haye been postulated>

Comprehensive examination of the large Nevertheless, strong "second returns" of

body of observations made during the past energy appear at about two-thirds of the

six years on the wave motion of the earth observing stations, in various locations at

at distances of 1 to 1000 km from individ- 100 to 150 km from the shot. These have

ual explosions has been emphasized dur- been tentatively identified as reflections

ing this year. Additional field observations from a sharp transition between the upper

are no doubt needed for the study of many crustal rocks (below the sediments), which

obscure points, but our purpose has not have average wave velocities of 6.0 or 6.1

I0 4 CARNEGIE INSTITUTION OF WASHINGTON

km/sec, and the underlying ultrabasic sharply refracted) ray should give an addi-

rocks of higher velocity (8.0 or 8.1 km/ tional set of data. Data showing the loca-

sec) which are believed to comprise the tions of appearance and disappearance of

outer parts of the mantle of the earth, these reflections make it possible to exclude

Nowhere have velocities as low as 5.6 certain variations of velocity versus depth,

km/sec (the old earthquake value for all For example, the velocities and crossover

"granitic" rocks) been found, nor is there points may be used to calculate the depth

any reliable evidence for intermediate ve- of crust and point of critical reflection on

locities of 6.3, 6.7, or 7.3, which have been the basis of a simple hypothesis of no ve-

traditionally assigned to the "intermediate locity change with depth. Critical reflec-

basaltic layers." Instead, we consider that tion points at 63=1=4 and 86 ±6 km are

the real fluctuations in our travel times predicted from the crossover points for the

show that the 6.0 and 8.1 velocities are Maryland and Minnesota regions, respec-

not specific values for specific types of tively. The observed points for first ap-

rock, but are averages in each case for the pearance of strong reflections are about 85

velocity of waves through many kilome- and 130 km, respectively. The discrepancy

ters of mixed rocks. There is no evidence may be most easily explained by the effect

that the 8.1 velocity for the mantle repre- of an increase in velocity with depth in the

sents anything but an average over many crust. Thus we are led to conclude that

kilometers, and it is entirely possible that the velocity of the crustal rock does in-

structures occur in these mantle rocks as crease with depth, and certain limits may

well as in the crustal ones above them. be set for the manner in which the veloc-

The observations show that the velocity ity changes with depth. For example,

of propagation in crustal rock must in- under Maryland the average velocity of

crease with depth, from 6 to say 7 km/sec. the upper two-thirds of the crust could be

This conclusion is based on additional in- 6 km/sec and that of the lower third

formation derived from the observation of slightly greater than 7 km/sec.
critical reflections. Without critical reflec- Furthermore, in the lower third of the

tions, there are usually only three variables crust, the velocity need not be constant,

which may be observed in a crustal seismic It may be constant at 7 km/sec or may

study; namely, the upper crustal velocity, start at 6 and increase ever more rapidly

the upper mantle velocity, and the distance with depth to 8 or just under 8 km/sec.

from the explosion at which one velocity More data will be necessary to choose

merges into the other (the crossover among the possible transitions in this

point). From these data it is possible to lower zone of the crust,
compute the depth at which the seismic In our search to establish the existence

velocity changes from the crustal velocity of the critical reflection and to discover the

(6 to 7 km/sec) to the upper mantle ve- qualitative nature of the crustal structure,

locity (about 8 km/sec). Now there is we have chosen to observe at points along

a large variety of possible velocity-versus- the earth's surface where we should most

depth relations which will fit these data likely encounter the strong second arrivals

equally well; and the usual seismic studies associated with total reflection (or refrac-

afford no criterion which singles out one tion). At present we are making addi-

such relation from the others. If, however, tional observations near the point of criti-

conditions are such that a strong reflection cal reflection. Data of this type should

may be observed, then this reflected (or give us more reliable information con-

DEPARTMENT OF TERRESTRIAL MAGNETISM

105

cerning the nature of the crustal transition.
There is no indication in any of our meas-
urements, at any depth, of a layer with
lower velocity than that of the rocks above
it, and we see no reason to think that such
a layer might be useful in explaining any
of our observations.

Several features of seismic observations
stood out as unexplained puzzles at the
beginning of the year. One of these is the
general phenomenon of reverberation:
Wave energy from a distant shot arrives
at an observing point; one might expect
the wave to pass on and the observing site
to regain its tranquility. It is a conspicuous
feature of all seismic observations that this
does not happen; after the "first arrival"
of waves, by whatever path may be appro-
priate to the position of the observing site,
the ground continues in strong unrest,
comparable to the disturbance when the
wave front first reaches the instrument.
Wave fronts may travel by several paths
through the crust and the mantle, and may
be of compressional or shear type, but they
are usually separated by many seconds at
the receiving site. The problem is to ac-
count for the shaking of the ground which
continues between these events. Theoreti-
cal analyses of the travel of waves in a
model homogeneous earth show no con-
tinued vibrations. It is conceivable that
if inhomogeneities and topographic fea-
tures of the order of a wave length were
introduced into these models, undulations
like those recorded in a seismograph might
result. Incidentally, reverberation has a
disastrous effect on "signal-to-noise ratio,"
as a wave packet arriving later by a slower
path cannot readily be observed on the
background of large unrest which remains
after the earlier wave arrivals. This phe-
nomenon of reverberation is related to two
or three other puzzling features, including
that of partial coherence in wave patterns
over many seconds at a set of receiving

stations spread out over as much as 2 km.
The motion of a given point on the surface
is of course necessarily related to the mo-
tion of adjacent points, but the apparent
velocity of these coherent wave patterns
traveling over the surface for many kilo-
meters corresponds to the velocity of the
deep-lying rocks (6.1 km/sec), and not to
the velocity of the soil underlying the
instruments (as low as 3 km/sec in some
cases) . That is, the waves may have the true
(group) velocity of secondary Rayleigh
waves, but an apparent (phase) velocity of
the originating wave. These observations
give rise to questions as to the strength or
even preponderance of local effects of ter-
rain near the observing sites. Such effects
might arise, for example, from conversion
of one wave type to another at a surface. If
local effects predominate, one must ques-
tion the reliability of the detailed motion of
the earth at any point on the surface as an
indicator of the waves arriving in that gen-
eral vicinity from the shot. This local sur-
face motion may be very strongly affected
by near-by topographic structures, and thus
it may be nearly impossible to get a repre-
sentative measure of the actual wave mo-
tion in the heavier rocks below the surface
by any instruments which can be placed
upon the surface.

Other questions concern the velocity
transition from 7 to 8 km, here referred
to as the "Mohorovicic discontinuity," fol-
lowing old terminology. Is this in fact a
discontinuity, giving rise to a reflected
wave, or may it be a transition zone sev-
eral kilometers thick, giving rise to a
strong second arrival by focusing effects
on refracted waves? What limits may our
data set on the possible variation of veloc-
ity with depth from the surface down to
the Mohorovicic discontinuity? A fur-
ther question concerns the degree of vol-
ume inhomogeneity or structure in the
crustal rocks, and the existence of struc-

I0 6 CARNEGIE INSTITUTION OF WASHINGTON

tures of varied rocks in the mantle region volume of the medium are converted near

below the Mohorovicic discontinuity. As the receiver site to surface waves, which

a part of the whole program, which is di- can spread out in only two dimensions in-

rected toward the elucidation of continental stead of three, a basis is given for under-

structures, their geological history, and the standing the continued ground motion

way in which they differ from ocean basins, after the arrival of the first wave fronts,

it is also appropriate to examine what other In this case the energy cannot be reflected

types of measurement, for example gravity back into the volume, but dies away with

variations, can give information on such time and distance much more slowly as

questions about crustal rocks. it moves away in the surface layers.

Several features stand out in the analy- This conception of the importance of
sis of our large series of seismic shot rec- conversion near the receiving site, which is
ords. One of these, as noted above, is the necessarily on the surface of the earth with
reverberation or continued ground motion all its topographic irregularities, led us to
which follows the arrival of any conspicu- undertake analytical experiments to ascer-
ous group of waves; another is the tan- tain whether either the first arrival or the
talizing degree of coherence or quasi peri- strong Mohorovicic reflection (both of
odicity of the waves as they pass over a which for brief intervals are clear-cut corn-
group of instruments spread over 2 km, as pressional wave impulses) is converted in
shown by our many tests with 24 and even a cycle or two into Rayleigh wave motions
48 simultaneous records using seismom- which continue as "reverberation." Theo-
eters 150 to 300 feet apart (pi. 3). Coherent retical work of H. Lamb (1904), H. Na-
wave groups were sought as reflections kano (1926), and E. R. Lapwood (1949)
from the horizontal intermediate layers at indicates that conversion to Rayleigh and
depths of 6 to 20 km which have so long more complex waves is expected when
been postulated in the earthquake litera- compressional waves encounter strongly
ture. The coherent wave patterns observed curved boundaries. Rayleigh waves give
over a considerable area at the surface characteristic elliptical motion to the par-
resemble the wave returns to be expected tides of an idealized surface. To examine
from such layers at intermediate depths, this hypothesis, a special combination of
but they are strongly dependent on shifts two strain seismometers and one vertical
of the shot position as well as on shifts of seismometer was devised, such that Ray-
the receiving array of seismometers, hence leigh waves impinging on the array from
they do not correspond to returns from any direction in the surface give rise to
buried horizontal boundaries. The strong zero net response, although the seismom-
dependence of these coherent wave group- eter array retains full sensitivity to com-
ings on receiving site, observed in Mary- pressional or other wave types. If the
land, Minnesota, and Washington, sug- reverberation consists chiefly of Rayleigh
gested to us that a possible phenomenon of waves from various directions in the sur-
major importance in all seismic observa- face, this new seismometer array will give
tions is the conversion of compressional a great enhancement of signal-to-noise ra-
waves which impinge on a surface, es- tio in detecting the arrival of successive
pecially an irregular surface, to surface new "phases," which are compressional or
waves of various types, especially Rayleigh shear waves arriving by various ray paths,
waves. If body waves arriving through the This could be of great practical importance

Department of Terrestrial Magnetism

Plate 3

ARRAY

TOP NORTH KEYS, MO. (NW) MARCH 6, 1951 A=47.7KM

LOWER ETHEL, VA. (SW) MARCH 15,1951 A=48.4KM

Multiple seismograms for studies of the earth's crust. The two records of explosions about 50 km
distant show the ground motion during 2.4 seconds after the first observed arrival (9 seconds after
the shot). Each set of 24 traces shows recordings of pairs of seismometers set 150 feet apart, ex-
tending 2400 feet in the direction of the shot and 1200 feet perpendicular to the shot direction. The
upper record was made 48 km northwest of the shot, and the lower 48 km southwest of the shot. The
two sets of seismograms show the great differences in detail of two similar situations and illustrate
the phenomenon of "reverberation."

DEPARTMENT OF TERRESTRIAL MAGNETISM

107

in geophysical prospecting. In addition, it
was realized that observations at a depth
of a wave length (about 2000 feet) or two
in a mine or borehole should show very lit-
tle reverberation if the latter phenomenon
consists primarily in the trapping of energy
in the form of waves which by their na-
ture are tied to the surface.

It is not always feasible to obtain ten or
twenty successive large explosions in ad-
vantageous locations. Deep mine observa-
tions, which seem to call for explosions in
the Great Lakes, have not yet been feasible.
The strain seismometer array for analyti-
cal tests seeking Rayleigh conversion waves
was ready for observations in January 1953,
but for several months thereafter the U. S.
Navy was unable to carry out its normal
program of test explosions in and near the
Chesapeake Bay. When observations could
be begun in May on relatively small ex-
plosions, it was found that the ground mo-
tion at several sites was clearly more com-
plicated than a simple Rayleigh wave
following the compressional phase. It is
surprisingly difficult to locate sites on ex-
posed rock surfaces which undergo appro-
priate strain as the shot waves pass; the
total ground motion, vertical and hori-
zontal, seems almost arbitrarily related to
the amount and direction of strain in the
rock face. These experiments make it evi-
dent that much of the surface of the earth
is like rubble on a large scale, and no
simple wave forms can be expected from
instrument arrays on the surface, even if a
simple wave impinges on such a surface
from below.

These experiments are actively continu-
ing as the report year ends, and they ap-
pear to demonstrate the very strong local
influence of contour and topography, in-
cluding buried topography, in the imme-
diate vicinity of a seismometer receiving
site. The strong conversion of body waves
to surface waves thus appears to be a

sound basis for understanding the local
coherence of waves over a region several
kilometers in extent; the propagation of
phase patterns which have the appearance
of a 6 km/sec velocity over surface sedi-
ments where the velocity is known to be
much lower; and the phenomenon of re-
verberation, by trapping wave energy near
the surface. Methods for signal-to-noise
reduction which are based on idealized pic-
tures assuming homogeneous media seem
inevitably to be inapplicable to the rough
surface of the real earth.

In order to be sure that, in our efforts
toward critical evaluation of seismic record-
ings, simplified but not erroneous deduc-
tions are made concerning the crust, we
have made numerous instrumental com-
parisons during recent years. It has been
important in field work to be able to use
filter discrimination against stray 60-cycle
pick-up, for example, and the relative
weakness of high-frequency components
at distances of 50 to 300 km from a shot
has seemed to justify elimination of high-
frequency response. During the autumn
of 1952 a series of tests were carried out
using our standard electromagnetic instru-
ments (velocity response) with frequency
range about 3 to 30 cycles, comparing these
with our older large seismometers (1 to
30 cps) and with a high-frequency ve-
locity instrument with photographic re-
cording (3 to 300 cps). In addition, a dis-
placement response instrument (capacity
bridge; response 1/3 to 30 cps) was oper-
ated at the same sites. All these instru-
ments made practically identical records,
and unequivocally the same interpretation
would have been reached using any one
of them. We therefore conclude that our
surveys have not been biased by instru-
mental response characteristics. On the
basis of these tests, several sets of compact
portable equipment, both two-position re-
connaissance units and six-position seismic

I0 8 CARNEGIE INSTITUTION OF WASHINGTON

"strings" for linear arrays (1600 feet), were (stability). Thus, the findings of the past

constructed and have been found very con- year put us in a better position ultimately to

venient and reliable for field use. select rocks whose magnetizations can be

used in tracing the history of the earth's

Rock Magnetism magnetic field in geologic time.

John W. Graham On the basis of our earlier observations,

At the beginning of the report year, a we had offered ( Year Book No - 5°. '950-
number of important questions remained '95 1 ) the suggestion that the direction of
in the broad problem of interpreting rock ma g n etization of some rocks, under par-
magnetism data. Whereas our studies on tlcular circumstances, could undergo a self-
sediments favor the view that the earth's reversaL Observations reported during
magnetic field has remained essentially J 95 2 h Y colleagues in the U. S. Geological
constant in direction throughout geologic Surve y and m J a P an a PP«red to confirm
time, the work of others on igneous rocks thls suggestion: Balsley and co-workers
appears to suggest that there may have re P orted a correlation of high titanium con-
been a number of reversals of the sense tent ln certain rocks wlth an inverse P olari -
of the field. Both these interpretations are zatl0n > and Na g ata sported on a rock,
subject to uncertainties because, first, there hl g h in tltamum > that squired an inverse
is only partial understanding of the mech- moment when it was cooled from high
anisms by which rocks become magnet- temperature m the earth's magnetic field,
ized, and, second, the factors that control These observations, together with Neel's
the stability of the magnetizations in time earller theoretical treatment of the prob-
have not been adequately established. lem > suggested to us a group of field ob-

,-. , .i • j- . . .. i iL servations which were expected to give

Our work this year was directed at both ,. . f °

,u ..• i .i i u j- direct confirmation of the hypothesis that

these questions, primarily through studies ,e

on igneous rocks. Out of this work has Certain 1 § neous ™ cks ™ th inverse ma §'

come the appreciation that magnetized negations are to be explained, not on the

minerals may develop in rocks at more basis of a °nce-reversed earth's field, but

than one time, for example by oxidation, rather on the basis o£ the P resen ce in the

and that the later-formed species may de- rock of two distinct ferromagnetic min-

rive their magnetizations from the pre- eral s P ecies havin S different Curie points,

viously magnetized material, rather than and an interaction of their magnetic fields,

directly from the earth's magnetic field. Jt was suggested that there might be

We have recognized that impurities con- found, on the margins of an igneous intru-

tained in magnetite at high temperature sive characterized by an inverse magnetiza-

may have an important influence on mag- tion, a zone of rapidly cooled rock having

netic properties at lower temperatures as normal polarizations, the thought being

they are rejected from the magnetite struc- that suitable rapid cooling would suppress

ture upon cooling, by exsolution processes, the development of the two phases necessary

The primary effect of eliminating impuri- for the self-reversal. This suggestion was

ties is in controlling the grain size of the put to test during six weeks of field work

magnetite. From magnetization experi- on a group of small and large Pre-Cam-

ments we now realize that grain size is brian diabase dikes in Michigan known to

important in controlling the ease with have inverse magnetizations. Approxi-

which the magnetization can be altered mately 250 selected samples were measured

DEPARTMENT OF TERRESTRIAL MAGNETISM

109

in the field, and provided a basis for fur-
ther study during the winter. Contrary to
expectation, samples from even the mar-
ginal fraction of an inch of a dike only a
foot and a half wide had the same sense of
magnetization as samples from the interior
of a related dike 150 feet wide. Since
abundant evidence is available that rapid
cooling did in fact take place on the mar-
gins, field confirmation of the self-reversal
mechanism was not obtained.

The laboratory work that followed made
apparent the reason for this result. By a
combination mainly of X-ray diffraction
measurements (made at the Geophysical
Laboratory) and heating experiments it
was demonstrated that the fraction re-
moved from the rock by simple magnetic
separation consists of a mixture of two
distinct ferromagnetic species — magnetite
(Fe 3 4 ) and maghemite (YFe 2 3 ) — inti-
mately intermingled with each other and
with "nonmagnetic" ilmenite (FeTi0 3 ).
Maghemite is not a primary constituent of
igneous rocks, but is developed at rela-
tively low temperatures by the oxidation of
magnetite. On being heated to 400-600° C,
it is converted to nonmagnetic hematite.
Hence, high-temperature experiments to
determine the original magnetic behavior
of these rocks cannot be made, because
part of the magnetic material of the natu-
ral sample is destroyed in the experiment.

During the past few years, solid-state
physicists have made great progress toward
an understanding of the magnetic prop-
erties of tiny particles. Their outlook has
been used to good advantage in some of
the problems of rock magnetism. For ex-
ample, by means of magnetization experi-
ments on the Michigan dike rocks, it has
been possible to demonstrate that the ferro-
magnetic ingredient behaves magnetically
like a mixture of small and large particles.
The small particles, whose state of mag-
netization is more difficult to change (high

coercive force), are responsible for pro-
ducing the magnetization observed in the
natural samples. The larger particles in
the natural samples are in a "demagnet-
ized" state, but may be easily remagnet-
ized. The combination of these two types
of particle in the rocks produces an inter-
esting magnetic behavior. If a sample is
subjected momentarily to a moderately
strong magnetic field (100 oersteds), its
state of magnetization is greatly changed.
If, however, this remagnetized sample is
now placed in an alternating magnetic
field of 100 oersteds peak value that is
allowed to decay slowly, the original state
of magnetization of the sample is restored.
Such behavior appears contrary to the
usual experience that a decaying alternating
magnetic field will demagnetize a sample,
but it may be easily interpreted in the light
of the recently acquired understanding of
the differences of behavior of small and
large particles. Microscopic observation of
this magnetic material indicates that the
magnetic properties are the consequence of
the macroscopic grains' being very finely
subdivided by the development of isolated
bodies of ilmenite which could not be re-
tained in solid solution within the mag-
netite crystal structure at low temperature.
These field and laboratory observations
have been synthesized in an internally con-
sistent hypothesis which explains how these
rocks could acquire their magnetizations
in an original magnetic field having a
sense opposed to that shown by the sam-
ples today. The magnetization mechanism
of these rocks is visualized as follows:
Magnetite, containing an abundance of
titanium in solid solution at high tempera-
ture, is minutely subdivided at lower tem-
peratures as excess titanium is rejected in
the mineral ilmenite; this complex type
of body becomes magnetized to high in-
tensity in the same sense as the applied
field as it cools below the Curie point of

CARNEGIE INSTITUTION OF WASHINGTON

magnetite; at still lower temperatures, oxi-
dizing reactions become important, and
maghemite is produced at the expense of
the more reactive small particles of mag-
netite. The net sense of magnetization of
these secondary maghemite particles is op-
posite to the sense of magnetization of the
primary magnetite and hence also oppo-
site to the sense of the magnetic field pres-
ent at the time the melt initially cooled.
With the passage of time, the larger, un-
altered particles of magnetite become de-
magnetized, and the sense of magnetiza-
tion of the sample becomes dominated by
the moments of the smaller particles — the
reverse of the original applied field.

Inasmuch as this magnetization process
requires many steps over a period of time,
it is possible that in nature the normal
sequence may be interrupted and later re-
sumed. In this way it would be possible
for the magnetic moment of sediments to
reverse at some time subsequent to deposi-
tion. Such appears to have been the case
in some of the sediments within the Ap-
palachian geosyncline, which were sub-
jected to elevated temperatures and deep
burial after their deposition.

Observations of the directional magnetic
susceptibility of samples from the Michi-
gan diabase dikes have indicated that
turbulent flow or movement below the
Curie point is not the cause of the scattered
directions of magnetization observed in
some exposures. Tentatively, the scatter is
attributed to local variations produced in
the chemical composition of the diabase by
thermal gradients at the time the dike was
injected, but the manner in which the scat-
ter is produced is not fully understood.

Other studies of the anisotropy of mag-
netic susceptibility made during the year
indicate that this property is related pri-
marily to textures produced in rocks at
the time of their formation or most recent
deformation. The subject warrants de-

tailed attention by geologists who are in-
terested in the textural and structural evo-
lution of rocks.

Isotope Dating of Ancient Igneous
Intrusives

L. T. Aldrich, G. L. Davis, L. O. Nicolaysen,
and G. R. Tilton

The co-operative program of staff mem-
bers of the Department of Terrestrial
Magnetism and the Geophysical Labora-
tory designed to ascertain age relations
between the igneous intrusives of Pre-
Cambrian time has been continued this
year. This program is a part of a general
focusing of attention here on questions
relating to the characterization of con-
tinental structures and their probable evo-
lution, with particular reference to the
long stretches of time before the Cambrian
period.

Until recently, methods of absolute age
determination have been limited in appli-
cation to ores of radioactive elements, prin-
cipally uranium and thorium, for which
gravimetric determination of the parent
and daughter elements is possible. During
the past few years an accurate method of
determining very minute quantities of ele-
ments having more than one stable isotope
has been developed, widening the scope of
age measurements to common rock-form-
ing minerals and to radioactive elements
other than uranium and thorium, such as
potassium and rubidium. The increased
sensitivity of this isotope dilution method
makes possible for the first time age deter-
minations by several radioactive decay sys-
tems on a single igneous rock mass, thus
providing an opportunity to compare the
reliability of the results obtained by the
different systems. When the reliability of a
method has been established, a large vari-
ety of geochemical and geophysical proc-
esses may be studied, including diffusion,

DEPARTMENT OF TERRESTRIAL MAGNETISM

III

rock alteration, and mineral formation.
The availability of a source of dated min-
erals makes possible a study of natural
nuclear physical rate processes such as
branching ratios, neutron flux in rocks, and
natural fission. At the beginning of the
present year's work, the rubidium-stron-
tium ages of five lepidolites and the ura-
nium-lead ages of the zircon and sphene
of the Essonville granite had been deter-
mined.

The activities of this year started with an
extensive field trip to the mapped Pre-
Cambrian areas of southern, southwestern,
and western United States. A party of
geologists, chemists, and physicists from
our two laboratories collected granites and
pegmatitic minerals for laboratory investi-
gation from selected localities in Texas,
New Mexico, Arizona, California, Mon-
tana, Wyoming, and Colorado. Wher-
ever possible, geologists working in the
particular areas were consulted. Their
support and assistance proved to be in-
valuable in obtaining representative ma-
terials.

Specimens were collected from sixteen
type localities of significant granite masses,
and lepidolites were found in five peg-
matites, four of which were seemingly as-
sociated with granite bodies. Accompany-
ing minerals such as feldspar, biotites, mus-
covite, amblygonite, beryl, and tourmaline
were also collected at the pegmatite locali-
ties for laboratory evaluation as to their
suitability either for age determination or
for measurement of isotope ratios. It is a
pleasure to acknowledge the assistance and
advice of the many people who helped
make the sample collection a success. They
include Dr. C. A. Anderson, Medora Krie-
ger, and Dr. T. S. Lovering, of the U. S.
Geological Survey; Dr. R. Jahns, Cali-
fornia Institute of Technology ; Dr. Arthur
Montgomery, Lafayette College; and offi-

cials of Grand Canyon, Yosemite, and
Yellowstone National Parks.

Thin sections of all rock samples were
prepared for petrographic examination,
and the granites were crushed for mineral
separation. Combinations of magnetic and
density separations were employed to yield
pure samples of the constituent minerals.
In some unfavorable cases 50 pounds of
rock yielded but 100 mg of zircon, barely
enough for preliminary analysis. Mineral
fractions of zircon, sphene, and biotite
were extracted from granite collected at
Cape Town, Union of South Africa, and
from a heavy mineral concentrate from a
granite in Southern Rhodesia. The latter
was obtained through the co-operation of
Dr. A. M. MacGregor, of the Rhodesian
Geological Survey.

Determination of Thorium by Isotope
Dilution

Considerable attention has been given
by authors of various papers in recent years
to quantitative chemical measurements
made by the method of isotope dilution.
The availability of stable isotopes from the
Atomic Energy Commission has made the
method practicable. An example of the
method, the determination of thorium, is
given here, since it demonstrates both the
method and a type of physical measure-
ment which can well be made on suitable
rocks.

Thorium in thorium-bearing minerals
has but one isotope, Th 232 , as shown in
figure \a. To determine thorium by the
method of isotope dilution, another iso-
tope of thorium in milligram quantities
must be found. Th 230 , not yet available
from the AEC, is a daughter of U 238 , of
sufficiently long half life to be present in
nearly pure form in originally thorium-
free uranium minerals. About 1 mg of
such thorium was extracted from an old
pitchblende and made into the carrier

112

CARNEGIE INSTITUTION OF WASHINGTON

solution for determining thorium in min-
erals. The isotopic composition of the car-
rier material is shown in figure \b, from
which it is seen that the ratio Trr 30 /Th
is about 4. The carrier solution is cali

232

thorium in a rock, a known volume of car-
rier solution is equilibrated with a solution
containing all the thorium of the speci-
men. A fraction of the thorium sufficient
for mass spectrometric analysis is then ex-

230 232
NORMAL THORIUM

230
230

232 230 232

CARRIER 3.496 y NORMAL Th

+
0.100 ml CARRIER
SOLUTION

230 232

0.100 ml CARRIER
SOLUTION

0.0946 g CAPE ZIRCON

Fig. 1. Mass spectra for the determination of very small amounts of thorium in mineral sam-
ples by isotope dilution, (a) Thorium alone (no Th 230 ). (b) Mass spectrum of Th 230 carrier,
Th 230 /Th 232 = 3.9. (c) Mass spectrum for calibration of carrier solution, 3.5 [ig normal thorium
+ 0.1 ml carrier solution. Th 230 /Th 232 = o.92, hence Th 230 content of carrier solution is 4.2 \xg/
0.1 ml. (d) Mass spectrum of thorium from mixture of 0.1 ml carrier solution and 0.09 gm zircon
from the Cape granite. Th 230 /Th 232 r=o.25, and Th 232 content of zircon = 15.7 (jg/0.0946 gm zircon.

brated from this ratio and from that of a
mixture of a known volume of the carrier
solution to a known mass of Th 232 . The
isotope distribution of the mixture is
shown in figure ic. The data for the
calibration of the carrier solution are now
complete, and as many minerals may be
analyzed as necessary. To determine the

tracted and analyzed, and another isotope
distribution is obtained (fig. id), from
which the thorium content of the rock or
mineral may be calculated. The method
is applicable to all elements which have
two stable or long-lived isotopes, and is
only slightly more complex when the ele-
ment to be analyzed has several isotopes.

DEPARTMENT OF TERRESTRIAL MAGNETISM

113

The application of the method to the
determination of Th 230 and U 238 in suit-
able uranium minerals will provide the
data required for calculating the ratio of
the decay constants of these two isotopes.
This extension of the method to the meas-
urement of nuclear constants promises to
be fruitful.

Elements which have been determined
by the method of isotope dilution at our
laboratory now include uranium, thorium,
lead, strontium, rubidium, and calcium.

Rubidium-Strontium Age
Determinations

The rubidium-strontium measurements
have been extended to include some 25
lithium micas, obtained from pegmatites
located in Pre-Cambrian areas of South
Africa and North America. Table 1 gives
the measurements on South African peg-
matites, and table 2 those for North Amer-
ica. The absolute ages are determined
from a decay constant for Rb 87 of 1.13X
io -11 yr _1 , the average of two recent deter-
minations by absolute beta counting.

The ages (in millions of years) presented
in table 1 confirm last year's findings
on the existence of minerals whose ages,
as measured by the rubidium-strontium
method, are greater than 3000 million
years. In addition, the ages of specimens
collected from different pegmatites in a
geologically homogeneous area agree with-
in 10 per cent. In one instance, however,
the disagreement between these measure-
ments and published syntheses of African
Pre-Cambrian tectonics is so striking that
the need for reconsidering this problem is
indicated.

The results in table 2 show for the first
time the existence of Pre-Cambrian pegma-
tites in the United States as ancient as any
in North America. The three pegmatites
with ages of about 3500 million years are

located roughly between the 2000-million-
year-old pegmatites of the Black Hills, re-
ported last year, and the 2700-million-year-
old pegmatite at Bagdad, Arizona. The
existence of very ancient minerals in the
United States as far south as New Mexico

TABLE 1

Indicated age of South African pegmatites

Age
Location (million

years)

Popes Claim, S. Rhodesia 374°

Letaba, Transvaal 3850

Hombolo, Tanganyika 3 2 5°

M'bale (1), Uganda 2460

M'bale (2), Uganda 2370

Lunya, Uganda 2 34°

Kinderzitt, Namaqualand 1210

Jakkalswater, Namaqualand 1100

Muika (1), Belgian Congo 1220

Muika (2), Belgian Congo 1210

Karibib-Usakos II, SW. Africa 600

Sahatany, Madagascar 690

Alto Ligonha, Mozambique 570

Dogon Daji, Nigeria 625

TABLE 2

Indicated age of North American pegmatites

Age
Location (million

years)

Bagdad, Arizona 2660 ±260

Bonneville, Wyoming 357° — 3 00

Dixon, New Mexico 3450 + 400

Ohio City, Colorado 344° — 3 00

will certainly contribute to theories of con-
tinent formation. These measurements are
also a part of the study of the correlation
of ages indicated by different radioactive
decay systems, which has the broad objec-
tive of determining the reliability of age
measurements by isotope ratios.

Large variation in the isotopic abun-
dances of the isotopes of strontium thought
to be stable was reported last year. An at-

ii 4

CARNEGIE INSTITUTION OF WASHINGTON

tempt was made this year to see whether
this variation could be due to radioac-
tive decay of zirconium or uranium con-
tained in the mineral zircon. The absolute
amount of strontium in the blank run in
parallel with the zircon was 2.5 ± 0.5 micro-
grams. That found in one gram of zircon
was within 20 per cent of this amount.
This can be considered good agreement be-
tween two blank runs. It is therefore con-
cluded that the radioactive decay of zir-
conium or uranium in the amounts present
in the mineral biotite could not contribute
in any way to the variation of the ratio
Sr 86 /Sr 88 found in biotite. It is of interest
to note that the strontium determination in
zircon was made using Sr 87 from lepidolite
as carrier material. The purity of this iso-
tope in the strontium in lepidolite is so
high (99 per cent) that it is ideal for the
determination of small amounts of com-
mon strontium.

Lead Age Program

Thorium-lead age determinations on the
mineral constituents of granites which are
the first to include isotopic analysis of the
lead have now been made. New uranium-
lead ages have been obtained for those min-
erals which had not been analyzed previ-
ously. The lead analyses are the collabora-
tive work of Dr. Claire C. Patterson, of
the California Institute of Technology.
Dr. M. G. Inghram, of Argonne National
Laboratories, generously provided mass
spectrometric facilities for the measure-
ments made during the first half of the
year. Granites from three localities have
been studied during the past year: Esson-
ville, Ontario (thorium ages) ; Uncom-
pahgre, Colorado (uranium and thorium
ages) ; and Cape Town, Union of South
Africa (uranium and thorium ages). The
Uncompahgre work is part of a detailed
study of the distribution of lead, uranium,

and thorium in the various mineral constit-
uents of the granite, and of the isotopic
composition of the leads, which will sup-
plement a similar study on the Essonville
granite already completed. The Cape gran-
ite measurements are part of the study of
thorium ages.

The determinations are summarized in
table 3. The ages given have an analytical

TABLE 3
Lead ages of granitic minerals

Sample

Age (million years)

U 238 /Pt> 206 Pb 207 /Pb 206 Th 232 /Pb 208

Essonville
granite:
Zircon .... 1030 1060 415

Sphene.... 920±80 1060±200 202

Uncompahgre
granite:
Apatite.... 1050 1810±160 200

Biotite 3200 1700 1180

Cape granite:

Zircon 347 550 278

accuracy of about 5 per cent unless other-
wise indicated. The Cape zircon ages are
based on preliminary data and are ac-
curate to about ±10 per cent.

Thorium Ages

The striking feature of the thorium-lead
ages is the fact that they are all much
lower than the uranium-lead and lead-lead
ages of the same minerals. This discrep-
ancy has been found by other workers for
a number of the uranium-thorium ores. In
particular, five of the six monazite age de-
terminations reported in the literature re-
semble the data of table 3 in having low
thorium-lead ages as compared with the
uranium-lead ages. No adequate explana-
tion has been offered for these observa-
tions. Since it now appears that many
granitic minerals can be expected to yield

DEPARTMENT OF TERRESTRIAL MAGNETISM

115

low thorium-lead ages, it is probable that
the goal of future thorium work will be
more to determine the cause of the dis-
crepancy than to measure ages by this
method. Some progress has been made
with the problem during the past year,
although the data are too few to permit
generalization.

The comparison between the Essonville
and Cape zircons is noteworthy because
X-ray examinations showed that the crys-
tal structure of the Essonville zircon is
badly shattered by radiation damage (met-
amict), whereas that of the Cape zircon
is nearly intact owing to its younger age
and lower radioelement concentration.
Since zircon is one of the most stable min-
erals known, the Cape zircon represents
one of the most favorable subjects for
thorium-lead age measurements on a rock
mineral. It is seen that the thorium-lead
and uranium-lead ages agree much more
closely for the Cape than for the Esson-
ville zircon, even allowing for the fact that
the Cape lead data are still of a prelim-
inary nature. This agreement indicates
that thorium age discrepancies in zircons
may be a function of the interrelated quan-
tities, the radioelement concentration, and
the extent of crystal damage.

An interesting result has been obtained
by leaching the Essonville sphene with
cold 6N hydrochloric acid and analyzing
the solution. The solution was found to
contain 20 per cent of the sample's ura-
nium, 40 per cent of the thorium, and 70
per cent of the lead. Moreover, the acid
removed 65 per cent of the radiogenic
Pb 208 , but only 25 per cent of the radio-
genic Pb 206 . Thus for this mineral the
lead generated by thorium is more sus-
ceptible to removal than that generated by
uranium. These observations would indi-
cate that loss of Pb 208 rather than addition
of thorium is the cause of the low thorium-

lead age for the mineral. It is possible to
lose Pb 208 without a proportionate loss of
Pb 206 if there is an inhomogeneous distri-
bution of uranium and thorium in the
crystals.

In the case of the Essonville zircon it
has been found that a hot aqua regia leach
will reduce the alpha activity of the min-
eral by 35 per cent. This leach solution
will be studied in the same manner as the
sphene leach solution, in an effort to obtain
further information on the problem of low
thorium-lead ages.

Uranium-Lead Ages

The uranium-lead age of the Cape zir-
con is less than the lead-lead age by an
amount that is far outside the limits of
analytical error. It is believed that the
lead-lead age is the more nearly correct
and that the uranium-lead age is low. The
Essonville zircon occurred in the rock as
large crystals several millimeters in di-
ameter, whereas the Cape zircon occurred
as small crystals but a few hundredths to a
tenth of a millimeter in diameter. The
hypothesis now favored is that the smaller
size of the Cape zircon crystals permitted
lead loss while the larger size of the Esson-
ville crystals prevented it. We have no
data for the diffusion rate of lead in zir-
con, but data exist for the diffusion of lead
in a perthite feldspar which indicate that
it could diffuse in the mineral over dis-
tances of the order of 0.1 mm in one bil-
lion years at 300° C under the influence
of a concentration gradient.

The Cape granite contains minerals
suitable for a rubidium-strontium age de-
termination, and such a study is planned
for the coming months.

The Uncompahgre measurements seem
only to show that the granite is older than
one billion years. It is not possible to state
at this time which of the age measurements

CARNEGIE INSTITUTION OF WASHINGTON

is likely to be the more accurate. A rubid-
ium-strontium age measurement on the
biotite will be made in the near future.

Transfer of Lead in the Essonville Granite

Further work on the Essonville granite
has given evidence of a transfer of lead
(and possibly uranium and thorium) with-
in the minerals of the rock. The perthite
present in the rock contains so much lead
as compared with uranium and thorium
that the isotopic composition of the lead
could not have been affected by radioactive
decay within the mineral during the bil-
lion years the granite has existed, and
should thus represent the lead present
when the rock was formed. When one
compares this measured isotopic composi-
tion with that calculated from the known
age and from uranium, thorium, and lead
data for the rock today, it is found that the
calculated ratios of Pb 206 , Pb 207 , and Pb 208
to Pb 204 are substantially lower (see table
4) . This indicates either addition of radio-
genic lead to the feldspar or else a frac-
tionation of lead from uranium and tho-
rium in the rock as a whole.

In view of this information, a large
perthite crystal from a pegmatite in the
Essonville area was analyzed for the iso-
topic composition of its lead. Diffusion
processes should have had much less effect
here, since the crystal was several centi-
meters in length, in contrast with the crys-
tals of the mineral in the granite, which
were but a few millimeters in length. The
pegmatite lead agrees very well with that
calculated from the total granite. It is
accordingly believed that there has been
no measurable fractionation of uranium,
thorium, and lead in the rock as a whole

and that there has been addition of radio-
genic lead to the feldspar since it was
formed. It is possible to leach highly radio-
genic lead (i.e., lead with high ratios of
Pb 206 , Pb 207 , and Pb 208 to Pb 204 ) from the
sphene of the granite with cold 6N hydro-
chloric acid. Such a process could serve as
the source of the lead which has been
added to the perthite.

If such transfers of atoms between the
various minerals of granites prove to be
the general case, the interpretation of age
measurements based on these minerals will

TABLE 4
Isotopic composition of Essonville leads

Sample 206/204 207/204 208/204

Perthite from granite

(observed) 18.4 15.6 39

Granite composite (cal-
culated) 16.4 15.3 32

Perthite from pegma-
tite (observed) 16.7 15.0 35

be greatly complicated. The studies of the
coming year should give an indication of
how important this factor will be.

The measurements made this year indi-
cate several things about the general prob-
lem of age determinations. The need for
cross correlation seems more important as
we accumulate data from one decay system
or another without comparison. The lead
measurements already indicate the diffi-
culty of determining unambiguously the
age of a granite from the ages found for
its mineral constituents. Determination of
the age by the rubidium-strontium method
may remove some of the ambiguity or in-
crease it. In any case, the granites already
on hand will provide the answer we need.

DEPARTMENT OF TERRESTRIAL MAGNETISM

117

THEORETICAL AND STATISTICAL GEOPHYSICS

Statistical Study of Seismic Waves
E. H. Vestine and S. E. Forbush

The seismic records of explosions pro-
vide geophysical time series which often
exhibit an unfavorable signal-to-noise ratio.
This sometimes happens for a compres-
sional wave reflection return from the
Mohorovicic layer at sites in the range 100
to 200 km from the shot point. When this
desired signal is weak, it may be lost in
the general background noise, and may
escape visual identification and timing in
the record. The question then arises
whether or not the desired signal can be
identified after the systematic noise is fil-
tered out by statistical methods. Similar
problems concerning a specific wave re-
turn or "phase" buried in the background
from waves arriving earlier comprise much
of the subject matter of seismic studies.
This general question in relation to geo-
physical prospecting was pointed out to us
by P. M. Hurley and his colleagues at the
Massachusetts Institute of Technology.

Statistical methods based on Wiener's
theory of stationary time series form a
convenient starting point in this type of
problem. By this means the record can be
analyzed into self-correlated (autocorre-
lated) time series and a random time
series. The wanted signal, supposed to be
of spiked form, can then be sought in the
latter time series by sampling techniques
applied to various intervals of record, us-
ing suitable significance tests.

These ideas were applied to three of
the velocity seismometer records obtained
by Tatel and colleagues from explosions
in Minnesota and in the Appalachian re-
gion. The particular records chosen were
ones for which Mohorovicic reflections
were clearly indicated by inspection of rec-
ords from other stations closer to the shot
point. In the records examined statistically,

a Mohorovicic reflection which escaped
visual detection was tentatively identified
as probable in one record of the three
examined. It was found that the autocor-
relation in the first 4 seconds of record was
the same or stationary, so that the auto-
correlated signal could be removed by pre-
diction technique.

It has been tentatively concluded that
the particular techniques used do not facili-
tate the identification of a wanted spike
signal in the noise unless the signal is large
enough to have been suspected on visual
examination of the record. Although based
on an analysis of only three records, this
result is not surprising and is entirely
reasonable. It is possible, however, that
cross-correlation techniques, suggested by
P. M. Hurley and his co-workers, involv-
ing a pair of suitably spaced recording sta-
tions, might have afforded some improve-
ment, especially for stations much closer to
the shot point.

In the case of a large blast in Tennessee,
the autocorrelation of the vertical velocity
component 1220 km from the blast showed
three remarkably stationary levels cor-
responding to the compressional, shear,
and Rayleigh wave phases, in the fre-
quency range of 3 to 30 cycles per second.
Rigorous statistical tests of the vertical and
horizontal seismometer records, however,
failed to show any persistent wave or ellip-
tical surface motion, corresponding to the
so-called Rayleigh wave phase, at these fre-
quencies. At 10 km from the Tennessee
blast, the autocorrelation was variable, in-
terval by interval in time, and thus unlike
that found at distances of about 200 to
1200 km.

An unsuccessful attempt was made to
identify the autocorrelated signal with sur-
face waves generated at the earth's sur-
face by reflected compressional waves, in

CARNEGIE INSTITUTION OF WASHINGTON

the manner suggested by H. Nakano and change in shape of the earth which might
E. R. Lapwood. It was concluded that the ensue and in turn give rise to a possible
available experimental data were of in- observed V 2 2 term in gravity,
adequate quality for this purpose, but suit- Several members of the Department
able analytical tests were developed which have been carrying on studies of the pos-
should facilitate further examination of sibility that toroidal magnetic fields due to
this question when satisfactory data on the electric currents generated at the juncture
space gradients of elastic displacements be- of mantle and core (thermoelectric cur-
come available. rents, for example) may in turn provide

The mechanics of wave propagation of Hall currents required to generate a sub-
the earth's crust is not well understood, stantial part of the main dipole and other
It seems likely from the studies already components of the earth's observed mag-
made that statistical and analytical methods netic field. Preliminary calculations are
based on modern signal-to-noise ratio sufficiently encouraging to show that this
techniques will be of material assistance in suggestion may merit serious consideration,
clarifying the nature of the wave propaga- This is because the field of a small dipole
tion involved. It may then be possible to inside the core in the presence of suitable
improve instrument design so that wanted toroidal electric currents will yield Hall
signals, when they occur, will be given currents effectively amplifying the dipole
greater relative prominence in the record, field. The Hall coefficient for the base of

the mantle is of course unknown, but no

Fluid Motions in the Earth's Core reason has yet been found for supposing

E. H. Vestine [t t0 be negligible.

At the invitation of the American Geo-
physical Union, a short study and discus-
sion of problems of the earth's central core
was undertaken. The evidence from the
fluctuating westward and northward plane-
tary motions of the geomagnetic field dis-
cussed in last year's report in relation to
fluctuations in the earth's rate of rotation
was reviewed. These motions of the geo-
magnetic field as a whole, noted at the
earth's surface, are expected to describe
corresponding motions in the surface re-
gion of the core. In the same way it should
be possible to determine from more local-
ized features of the changing geomagnetic
field some of the grosser motions occur-
ring locally in the surface region of the
core. A preliminary study suggests the
possible presence of two rising and two
falling centers of fluid motion near the
equatorial plane of the core. An attempt is
under way to check this by calculating the

Toroidal Magnetic Fields in the
Atmosphere

E. H. Vestine

A further point noted was that it may
be possible to check the generation of
toroidal magnetic fields on a geophysical
scale by means of rocket-borne magnetom-
eters in the upper atmosphere. In the
highly conducting auroral regions, varia-
tions in zonal wind velocity with height
are likely to generate large toroidal mag-
netic fields difficult to detect at the earth's
surface. Such fields may afford a terres-
trial explanation for effects on the latitude
distribution of cosmic rays, for part of the
variation with longitude in cosmic-ray in-
creases during large solar flares, and for
the existence and azimuth of homogeneous
auroral arcs. In any event, the conclusion
that toroidal magnetic fields exist in the
upper atmosphere seems inescapable. The

DEPARTMENT OF TERRESTRIAL MAGNETISM

II 9

main uncertainty is one of degree, namely,
whether any toroidal field comparable in
magnitude with the earth's observed mag-
netic field coexists in the atmosphere. This
appears to depend on the available electric
conductivity of the upper atmosphere, since
increases in wind velocity with height seem
assured on both theoretical and experi-
mental grounds. In this connection it
should also be noted that suitable winds
can tend to neutralize the Hall currents
accompanying toroidal electric currents in
the atmosphere. As T. G. Cowling showed
in 1932, under these conditions the effective
conductivity transverse to a magnetic field
in an ionized gas can approach that along
the magnetic field. This would give
enough conductivity to indicate closed-cir-
cuit toroidal fields several or many gauss
in strength.

Daily Mean Meridional Winds
E. H. Vestine

During the past seventy years it has from
time to time been suggested that winds in
the upper atmosphere generate the time
fluctuations of geomagnetism (including
those of magnetic storms) by dynamo
action.

The essence of this idea is that the mo-
tions of electrically conducting air across
the lines of force of the geomagnetic field
generate electric currents in the upper at-
mosphere. The changing electric currents
overhead then give rise to a fluctuating
magnetic field at the ground. A number
of studies have shown that the ordinary
periodic daily change in the geomagnetic
field is satisfactorily explained by this
mechanism. But it has never been shown
that the much larger occasional changes in
field known as magnetic storms are to be
explained in terms of probable vigorous
motions of the upper air, although O. R.
Wulf in a series of papers has recently re-

newed considerable interest in this pos-
sibility. An alternate incomplete theory of
magnetic storms ascribes practically all ef-
fects to solar streams of charged particles
as a primary cause. According to this
theory the initial phase of storms is due to
electric currents flowing within the solar
stream at a distance of a few earth radii.
This current reverses and becomes stronger
during the main phase, with leakage of
charge to polar regions yielding electric
driving forces for electric currents flow-
ing in the atmosphere.

A beginning was made in testing the
dynamo theory of magnetic storms. From
a study of upwards of 120 magnetic storms,
it was found that about 75 per cent of field
changes observed at the ground must be
ascribed to electric currents in the atmos-
phere, except possibly the part depending
on universal time, because the observed
changes at stations a few hundred kilo-
meters apart are too great, in all latitudes,
to be explained by electric currents flowing
at distances several earth radii away. In
co-operation with S. Chapman, it was
noted that this condition is also effective
during the initial phase of storms. Ac-
cordingly, only a small fraction of the ob-
served field pattern can reasonably be
ascribed to electric currents flowing be-
yond the atmosphere during a number of
the hours of storm.

As a more direct approach, an approxi-
mate meridional wind system for monthly
means of days of magnetic storm was de-
rived. Such a wind system cannot be
uniquely inferred from the geomagnetic
data. A simple derived system of winds
comprises rising air near the equator
which flows poleward, then returns at
lower atmospheric levels. It was found to
resemble closely the system obtained from
meteorology by W. W. Kellogg and G. F.
Schilling, even to the detail of a seasonal
term. The specific electric conductivity

I20 CARNEGIE INSTITUTION OF WASHINGTON

across the earth's magnetic field, estimated mental Station for the past year, during
from ionospheric data for the F region, which time no solar-flare increases have
however, is probably only about io~ 14 emu, been observed. A second meter with par-
so that improbably great wind velocities aflin-lead pile, which should be twice as
might be required at particular hours of a effective and much less heavy than the
great magnetic storm. It is possible to in- first, has been constructed. Tests have
crease the effective transverse conductivity been made which show that a reflecting
for the required westward-flowing cur- electrometer and a photomultiplier can be
rents by means of eastward-flowing zonal used to prevent the recording spot from
winds throughout a layer. Calculations in going off the film if neutron increases
progress suggest that the poleward-di- greater than about 300 per cent occur,
rected electric field generated by eastward- Though this type of continuously record-
flowing zonal winds will be used in gen- ing neutron meter has advantages of sim-
erating toroidal electric currents, but if plicity, the demands on the insulation are
some degree of north-south polarization high. So far it has not been found possible
results, the effective transverse conductiv- either to eliminate leakage loss or to keep
ity for westward current flow may be the loss constant for a period as long as a
adequate. month or two. Tests are under way with
Using a technique devised by J. A. Rat- a second counter to determine whether its
cliffe, estimates were made of total ioniza- insulation characteristics can be made satis-
tion of the F layers for Watheroo during factory.

night hours of magnetic storm. It was Old cosmic-ray program. Continuous

found that the average total ionization in operation of Carnegie Institution of Wash-

the F region probably changes by less than ington Compton-Bennett meters was car-

a factor of 2 during storm hours. This ried on at Godhavn (Greenland), Chel-

conclusion is possibly also supported by tenham (Maryland), Huancayo (Peru),

the fact that the average daily range in and Christchurch (New Zealand). Com-

the quiet-day diurnal variation is the same plete reduction of bi-hourly values of the

on days before and following solar flares, data from Cheltenham and Huancayo was

and also on days preceding and during maintained.

magnetic storms, so that the total ioniza- Secular change in burst frequency. The

tion of the E region changes little during measurement of bursts from the Compton-

storms. Consequently, any sources of ioni- Bennett records at Huancayo has been

zation, such as ultraviolet, which would continued. Results of careful statistical

increase conductivity in regions occupied analyses indicate that the burst frequency

by the supposed electric currents and cause has chan £ ed b Y about 6o P er cent in the

magnetic storms would have to be such interval *93fi-*95* and that the variation

that E and F regions are little affected * us £ar w compatible with a 22-year cycle.

1 r 1 1 • . .1 1 If this cyclic variation is substantiated by

berore and during storms, in low and r / ... ,. . ;

Mi . 1 future data, it would indicate an important
e latitudes. . ' , . , r

solar influence on the very high-energy part

r^ r> t> of cosmic rays conceivably arising from

Cosmic-Kay Research . ; . J rr* •

a changing solar-magnetic moment. This

S. E. Forbush | 3Urst mte a pp ears t0 De unaffected by

Neutron meters. One neutron meter has magnetic storms and to be unrelated to

been in operation at the Derwood Experi- sunspots.

DEPARTMENT OF TERRESTRIAL MAGNETISM

121

LABORATORY PHYSICS

NUCLEAR PHYSICS

N. P. Heydenburg, G. M. Temmer, and
Student Collaborators

The identification and complete specifi-
cation of nuclear energy levels continues to
be of considerable interest in the so-called
"low-energy" region of nuclear physics.
Once the energy and width, total angular
momentum, parity, and possibly the iso-
baric spin are determined for all energy
levels of a nuclear species, that species is
completely described and any successful
nuclear model must fit into this empiri-
cally established framework. The concept
of "isobaric spin" just mentioned has
moved into prominence in the past year
because of the striking similarities of the
level structures of mirror nuclei (indi-
cating charge symmetry, i.e. equality of
neutron-neutron forces and proton-proton
forces), and even of triplets of isobars (in-
dicating charge independence, i.e. equality
of neutron-proton forces and proton-pro-
ton forces). Corresponding levels are con-
sidered members of a "multiplet," that is,
different orientations of a spinlike quan-
tity; hence the term isobaric spin.

We have been making use of most of
the methods available for the determina-
tion of energy levels, such as elastic and
inelastic scattering of charged particles,
yields of various reaction products as a
function of the accurately known incident
energy, and angular distributions of neu-
trons, protons, and gamma radiation com-
ing from levels in the compound nucleus
under study. Because of the availability of
large currents of helium ions, we have con-
centrated largely on using alpha particles
as projectiles. Regions of excitation in the
compound nuclei thus become accessible
which can usually not be reached by other
means. Furthermore, the fact that alpha
particles have zero spin greatly simplifies

the assignment of energy-level properties
of the type discussed above.

During the past year we have continued
with our investigation of alpha scattering
and alpha-induced disintegrations. Alpha
scattering on nitrogen has been investi-
gated, and the work on alpha-alpha scat-
tering has been continued. Disintegration
experiments with alpha particles have been
done on nitrogen with the emission of pro-
tons and on carbon of mass 13 with the
emission of neutrons, and work has been
continued on the disintegration of beryl-
lium with the emission of neutrons. From
these experiments, new information has
been obtained on the energy-level schemes
of the compound nuclei formed in each
case.

In our report for the year 1951-1952
(Year Book No. 51), some results on alpha-
alpha scattering were discussed. Summa-
rizing these results, (1) it was found that
the scattering cross sections for alpha ener-
gies below 400 kev were in agreement with
the theoretical calculations of Mott, assum-
ing only Coulomb forces between the alpha
particles; (2) no anomaly in the scattering
was observed at the energy corresponding
to the ground state of the beryllium nu-
cleus of mass 8; (3) the effect of nuclear
forces between alpha particles became in-
creasingly apparent as the energy increased
above 400 kev. The data for energies up
to 3 Mev were analyzed in terms of S-wave
phase shifts; the phase angle varied from
180 at 400 kev to approximately 135 at
3 Mev.

During the past year we have made fur-
ther observations on alpha-alpha scattering
in order to determine whether the analysis
of our data in terms of S-wave phase
shifts was adequate, or whether phase
shifts of higher angular momentum were
also needed. D-wave phase shifts might be

I22 CARNEGIE INSTITUTION OF WASHINGTON

expected to be present even at 3 Mev (1.5 be desirable to follow this trend to higher
Mev center of mass) if the broad first ex- energies than we had available,
cited state of Be 8 at 2.94 Mev has a spin We have also extended our search for
value / =2. an anomaly in the alpha-alpha scattering,
Since the effect of a D wave vanishes at due to the effect of the ground state of Be 8 ,
the scattering angle = 27°22', it should to higher energies in order to cover the
be possible to determine a unique value for somewhat higher generally accepted value
the S-wave phase shift at this angle, if for the ground level of Be 8 (94.5 ±1.5 kev).
absolute scattering cross sections are known Still no effect has been observed. This
and if still higher-order phase shifts are enables us to set a lower limit to the life-
negligible. Measurements at other angles time of the Be 8 ground state of 2X10" 17
can then be used to determine any D-wave sec. It is known to be shorter than io" 15
contribution. Absolute cross sections were sec from cosmic-ray evidence,
measured by calibrating the scattering Continuing our program of study of
chamber with alpha scattering in argon, alpha-particle-induced reactions, we have
which was found to vary with energy and investigated the elastic scattering of alpha
angle according to the Rutherford scat- particles by nitrogen and the reaction
tering law over the energy region and N 14 + alpha -» F 18 -> O 17 '+ proton,
angles of interest in alpha-alpha scatter- which leads to the same compound nu-
ing. The alpha-beam currents were meas- cleus, F 18 , as the scattering. Only protons
ured in an evacuated Faraday chamber leading to the ground state of O 17 were
separated from the scattering volume by observed. By measuring the cross section
a 2 X io _5 -inch nickel foil. A phase-shift for scattering at large angles and the yield
analyzer was designed and constructed of protons, as a function of the energy of
which could be set up to give simulta- the incident alpha particles, we have lo-
neously combinations of S and D phase cated two new energy levels in F 18 .
angles giving the observed scattering cross At a scattering angle of 156° 20' in the
section at a given angle. A set of curves center-of-mass co-ordinates, rather sharp
was then plotted, with one phase angle as pronounced dips occurred in the number
the ordinate and the other as the abscissa, of scattered alphas at 2.88 Mev and at 3.05
one for each scattering angle. If there is a Mev, followed by a sharp rise after the
common intersection of these curves, a second dip, then a gradual return toward
unique value of S and D phase shifts is the yield expected from Rutherford's scat-
determined. At 2 Mev and below, only S tering law. The shape of these dips indi-
phase shifts are needed to fit our data. At cated that the resonances were sufficiently
2.5 and 3.0 Mev both S and D phase shifts broad and close together to cause inter-
are necessary. The S phase angle varies ference effects which made it difficult to
smoothly from 180 at 400 kev to 125 at determine the true resonance energies. The
3.0 Mev. A D phase shift of — 3 gave proton yield from the alpha-proton reaction
the best fit to the 3-Mev curve, although on nitrogen gave marked resonance peaks
the difference between the experimental at close to the same energies as the dips in
and theoretical curves at S = 2o° and 30° the scattering. From this curve the reso-
was greater than our estimated experimen- nance energies were found to be at 2.935
tal errors. This may indicate the presence and 3.140 Mev, corresponding to levels in
of still higher-order phase shifts. It would F 18 at 6.69 and 6.85 Mev, respectively. The

DEPARTMENT OF TERRESTRIAL MAGNETISM

I2 3

proton yield also indicated the presence of
a much broader level at 3.5 Mev or above.
The location of these levels in F 18 is shown
in figure 2, also the proton-yield curve and
the alpha-scattering cross-section curve.

In principle it is possible to determine
the spin and parity which characterize

and preliminary attempts at analysis did
not yield unique values for the spin and
parity.

A preliminary report was given last year
on the reaction Be 9 + alpha — > C 13 — >
C 12 + neutron + gamma. In this reaction
neutrons can be emitted, leaving C 12 in its

2m„C-

- 1.671 J / fi*0-64
ie

ENERGY LEVELS IN

ENERGY LEVELS IN 0'

Fig. 2. Schematic diagrams of energy-level systems of F 18 , C 13 , and O 17 investigated during the
year. The observed yield curves shown have led to the discovery of new energy levels. The left-
hand figure shows the results of the scattering of alpha particles on N 14 and the reaction:
N 14 + a — ■> F 18 — > O 17 + proton. From these curves new energy levels in F 18 have been assigned
at 6.69 and 6.85 Mev. The middle figure shows the yield curves for gamma and neutron emission
from the reaction Be 9 + a — ■> C 13 — > C 12 + neutron + y. The resonance peaks in these curves lead
to five new resonances in the nucleus C 13 . Also illustrated here are the two modes of disintegration
of the excited C 13 nucleus, either by neutron emission to the excited state of C 12 then to the ground
state by gamma emission, or directly to the ground state by emission of neutrons of higher energy.
The right-hand figure shows the energy levels for the nucleus O 17 which have been assigned from
the resonances appearing in the yield curve for neutrons from the reaction C 13 + a — > O 17 *->
O 16 + neutron.

these levels by observations on the scatter-
ing at a number of angles. Observations
were made at angles 90°, 125 16', 137
43', and 156° 20' (center of mass), as well
as several angular distributions at critical
energies. Because the nucleus N 14 has spin,
/=i in its ground state, and because the
observed levels are broad and close to-
gether, the theoretical analysis is complex

ground state or in its first excited state;
C 12 excited then goes to the ground state
by gamma-ray emission. Energy levels in
the compound nucleus C 13 were reported
at 1.92 and 2.62 Mev from observations on
the neutron and gamma-ray yields as a
function of the incident alpha energy.
More careful observations of the neutron
yields at both o° and 90° with respect to

124

CARNEGIE INSTITUTION OF WASHINGTON

the incident alpha beam indicated reso-
nances at 1.55, 2.27, and 2.84 Mev for neu-
trons which were not observed for gamma
rays. Hence these resonances appeared to
be excited only for neutron emission to
the ground state of C 12 . The energy-level
scheme for C 13 is shown in figure 2, to-
gether with the yield curves for neutrons
and gamma rays.

Studies have also been made on the re-
action C 13 + alpha -> O 17 -» O 16 + neu-
tron. The yield curve for neutrons from
thin targets, enriched 23 per cent in C 13 ,
gave resonances at the following alpha en-
ergies: 1.40, 2.23, 2.32, 2.45, 2.57, 2.68, 2.80,
3.13, and 3.36 Mev. These resonances cor-
respond to energy levels in the compound
nucleus O 17 at 7.41, 8.04, 8.11, 8.21, 8.30,
8.39, 8.48, 8.74, and 8.91 Mev, respectively.
The energy-level scheme for O 17 and the
observed neutron yield are shown in
figure 2.

During the year the Department's cyclo-
tron has been undergoing some alterations.
The magnetic field had never been prop-
erly shimmed for extracting the beam
from the circular path inside the dees.
Careful field measurements have been
made, and a rim shim has been added to
keep the field up near the edge of the poles
and to produce a strong fall-off field at the
proper radius. Pyramidal shims have also
been added to produce the proper decrease
of field from the center for proper focusing
of the beam as it spirals out from the cen-
ter. By the end of this report year the
cyclotron had been reassembled, and pre-
liminary measurements indicate an im-
provement in the amount of circulating
deuteron beams at large radii.

BIOPHYSICS

P. H. Abelson, E. T. Bolton, R. J. Britten,
D. B. Cowie, and R. B. Roberts

Six years of work in biophysics have
only increased our conviction that many

advantages accrue from the investigation
of biological problems in a physics depart-
ment. The points of interest in the field
of biology are somewhat different for the
physicist from what they are for the biolo-
gist, and the approach to a given problem
may be different. Even after some years
in the field, the physicist is still a "for-
eigner" and observes certain peculiarities
which are too familiar for the "local citi-
zens" to notice. Also, the biological work,
currently a quantitative study of "flow dia-
grams" for various chemical elements in-
volved in the biosynthesis of proteins and
nucleic acid, gains perspective when con-
stantly viewed in comparison with the
other problems under study in the Depart-
ment.

The interchange of ideas and techniques
is not a one-way process. The biophysics
group has aided the study of the age of
rocks by helping to initiate the use of ion-
exchange columns in mineral separation
for work at the microgram level needed in
the radioactive dating of very ancient igne-
ous rocks. It has also stimulated an ex-
amination of ancient rocks for organic re-
mains. We believe its main contribution
to the rest of the Department, however,
lies in keeping them aware of the nature
and complexity and elegant regularity of
processes in living matter, which cannot
fail to broaden their view of the universe.

A specific program in biophysics was
initiated at the Department in 1946 as the
outcome of a long period of collaboration
with biologists in the use of radioactive
tracers. For the past four years, the group
has directed its efforts toward studies
which might increase understanding of
the biosynthesis of proteins and nucleic
acids. This central problem of biology can
be attacked at many levels and by the use
of many techniques. One particular ap-
proach, a study of the biosynthesis of the
amino acids and other substructures from

DEPARTMENT OF TERRESTRIAL MAGNETISM

125

which proteins and nucleic acids are as-
sembled, has received the major emphasis
for some time.

During the past year the knowledge
previously gained was amplified and con-
firmed by extensive use of various C 14 -
labeled compounds. The pathways deter-
mined with C 14 2 as a tracer were also
demonstrated with C 14 -labeled acetate and
C 14 -glucose. The use of many different
tracer compounds also permitted quantita-
tive determinations of the flow through
the different reaction sequences of biosyn-
thesis. As a result of this work, it is now
possible to account quantitatively for most
of the carbon which the growing cells of
Escherichia coli convert from glucose into
cellular material.

The knowledge gained in these studies
of the biosynthesis of relatively simple
molecules can be applied in many ways to
the problems of the synthesis of the com-
plex macromolecules of protein and nucleic
acid. These large molecules are assembled
from their simpler components, which are
supplied in the required quantities by ap-
parently independent systems. One ques-
tion which now appears to be open to ex-
perimental study is the nature of the
controlling mechanisms which co-ordinate
the rates at which the needed materials
are formed. An approach to this problem
has already been made by examining the
material lost from growing cells. This
work has demonstrated that normally a
very precise balance between synthesis and
utilization is achieved by the cell and no
building blocks are synthesized in excess
of the requirements. Under certain condi-
tions this balance can be upset and large
quantities of various materials are synthe-
sized only to be excreted into the medium.

Another type of study utilizing this
knowledge of the synthetic pathways has
indicated the importance of peptides in
protein synthesis. Kinetic measurements

seem possible which would determine the
extent to which peptides serve as inter-
mediates. It appears more profitable to
proceed in these directions than to attempt
to work out the remaining puzzles of
amino acid synthesis.

The wide variety of experiments which
have been carried out could not be re-
ported adequately in the usual technical
journals. Consequently, it has been neces-
sary to devote a large part of the year's
work to assembling the information in
the form of a monograph which will be
submitted to the Carnegie Institution for
publication. This book will mark the
climax of one phase of the biophysics
program.

Amino Acid Synthesis in
Escherichia coli

The isotopic competition method has
been used with a variety of C 14 -labeled
molecules to determine the pathways of
synthesis for fifteen of the nineteen amino
acids found in the proteins of Escherichia
coli. This method consists in making more
than one source of carbon available to the
cell, and labeling one of the carbon sources
with C 14 so that the way in which it is
utilized by the cell can be determined. It
is frequently observed that the cell selects
one of the carbon sources for certain spe-
cific biosynthesis, using other carbon for
the rest of the cell material.

As an example, the amino acids threo-
nine, isoleucine, methionine, lysine, and
diaminopimelic acid all derive the same
quantity of carbon per molecule from CO2
as does aspartic acid. Consequently, they
may be grouped as the "aspartic family."
Furthermore, the location of the carbon
derived from CO2 corresponds throughout
the family. The same relationships are
shown when C 14 -glucose, C 14 -acetate, or
C 14 -aspartic acid is used as tracer. No one

126

CARNEGIE INSTITUTION OF WASHINGTON

of the family except aspartic acid, however,
gives rise to all the others, a fact which
shows that aspartic acid serves as the par-
ent substance for the whole group.

C 14 -labeled glucose, acetate, carbon di-
oxide, and amino acids have all been used
as tracers in these studies. Consequently,
the pathways have been established and
checked by several different methods. In

KC-^WVSPARTIC ACID-

S' 4 ''y * NUCL

EIC ACID, PYRIMIDINES
4.1

-►ASPARTIC ACID
11.5

NH 3

■►LYSINE
5.6

-4>— »-DIAMINOPIMELIC ACID
kL 0-4

HOMOSERINE

-♦-HOMOCYSTINE— -^METHIONINE

■ 3 2

CH 3 3Z

THREONINE

NH 3 k.NH 3

aNH-> BUTYRIC ACID * » <X KETO BUTYRIC ACID

Cp Wot

-♦THREONINE
5.0

KETO /3 METHYL VALERIC ACID

NH,

}

ISOLEUCINE
5.0

Fig. 3. Aspartic acid is formed as a product
of the Krebs cycle and is used by the cell in the
synthesis of a number of other products. More
than 50 per cent of the carbon utilized in pro-
tein synthesis is found in aspartic acid, glutamic
acid, and their products. The numbers indicate
the flow as observed in E. coli (micromoles per
gram dry cells per 100 seconds).

addition, glucose labeled with C 14 in the
1 or 2 carbon position has given additional
information on the pathways of carbohy-
drate metabolism. C 14 -formate and C 14 -
formaldehyde were also used to a limited
extent in examining the formation of
nucleic acids.

The sequence of reactions giving rise to
the aspartic family is shown in figure 3.
The presence of homoserine as an inter-
mediate was demonstrated both by the
competition method (see p. 130) and by

observing the production of radioactive
homoserine (from C 14 -glucose) when C 12 -
homoserine was added to the medium. As-
partic acid also contributes to the synthe-
sis of pyrimidines, thereby linking amino
acid and nucleic acid synthesis.

KC— =^-»-GLUTAMIC ACID \ » GLUTAMIC ACID
I 11.8

•PEPTIDES (GLUTATHIONE)

ACETATE

•N ACETYLGLUTAMIC
H 3-

ACETYLORNI

ACETATE
THINE— «^-*- ORNITHINE •

CITRULUNE-jr-*-ARGININE

CO,

-tt-*-ARGINI
' 4.9

Fig. 4. Glutamic acid, a product of the Krebs
cycle, is transformed by a series of reactions
to form proline and arginine. The numbers in-
dicate the flow into the various products in
micromoles per gram dry cells per 100 seconds
as observed with exponentially growing E. coli,
using radioactive carbon-14.

♦-NUCLEIC ACID
PURINES

GLUTATHIONE

Fig. 5. Serine is the precursor of glycine and
cystine and is also utilized in lipid synthesis (not
shown). The glycine is used in part for nucleic
acid synthesis; serine thus acts as a common
precursor of proteins, nucleic acids, and lipids.
The flows indicated are in micromoles per gram
dry cells per 100 seconds as observed in exponen-
tially growing E. coli, using carbon-14 as a radio-
active tracer.

Glutamic acid gives rise to proline and
arginine, as shown in figure 4. It is also
used in the formation of peptides, particu-
larly glutathione, which is abundant in
the cells.

Serine is the precursor of glycine and
cystine, as shown in figure 5. The glycine
and cystine thus formed are utilized in the
formation of glutathione. Glycine is also

DEPARTMENT OF TERRESTRIAL MAGNETISM

127

used in the synthesis of purines, and thus
it provides a second link between proteins
and nucleic acids.

Pyruvic acid is utilized in the synthesis
of alanine, valine, and leucine, as indicated
in figure 6. The conversion of valine to
leucine apparently proceeds by decarboxy-
lation and addition of acetate. In addition
to these major pathways of amino acid

scheme, glucose is converted to pyruvic
acid by way of a series of reactions involv-
ing phosphorylated intermediates. The
pyruvic acid then enters the Krebs cycle
(fig. 7) and is oxidized to CO2.

Although the operation of the Krebs
cycle in bacteria has been disputed, it

PYRUVIC ACID

CO?

y * ■ VALINE <£ -^LEUCINE

f 5.9 7.9

TO MEDIUM

LIPIDS

Fig. 6. Pyruvic acid plays an important role
in the synthetic activities of cells. The flow of
pyruvic acid to form alanine, valine, and leucine
and other products is given in micromoles per
gram dry cells per 100 seconds as observed in
exponentially growing E. coli.

synthesis, alternate pathways have been
shown to exist. In the usual medium these
pathways are unimportant, contributing
only a few per cent of the material. In
certain other media, however, they assume
major importance. The existence of these
potentially available mechanisms might
cause very misleading interpretations of
mutant requirements.

These amino acids account for 90 per
cent of the protein carbon or 54 per cent of
the total carbon incorporated by the cell.
Consequently, it was of considerable in-
terest to relate the quantities of amino acid
formed to the total utilization of glucose.

For some time the degradation of glu-
cose to CO2 and water has been believed
to proceed by two mechanisms working in
series. In the Embden-MeyerhofT glycolytic

Fig. 7. In the Krebs tricarboxylic acid cycle,
C0 2 condenses with pyruvic acid to form oxalo-
acetic acid. A portion of the oxaloacetic acid is
converted to aspartic acid, partly by transamina-
tion and partly by animation of fumaric acid.
Another portion of the oxaloacetic acid condenses
with acetic acid to form citric acid. A further
sequence of reactions converts the citric acid to
a keto glutaric acid, which serves as the pre-
cursor of glutamic acid. In addition, a part of
the a keto glutaric acid is decarboxylated to yield
the dicarboxylic acids, thereby regenerating ox-
aloacetic acid. The carbon atoms are lettered to
facilitate identification: the input carbons are
designated A, B, C, D, E, F, and the carbons of
aspartic acid are labeled W, X, Y, Z. When the
flow in the various branches of the cycle is
known, the radioactivity of the W, X, Y, Z car-
bons can be calculated from the known radioac-
tivity of the input carbons.

seemed a plausible mechanism for the in-
corporation of C0 2 and acetate into the
glutamic and aspartic families of amino
acids. It was found that predictions based
on the Krebs cycle were checked precisely
by the observed incorporation of various
tracers into the amino acids. Furthermore,

128

CARNEGIE INSTITUTION OF WASHINGTON

measurements of the incorporation of one
tracer (C 14 2 ) specified very definite flows
in the cycle which could be verified inde-
pendently by measuring the incorporation
of other tracer compounds. Figure 8 shows
a typical flow pattern for cells growing ex-
ponentially with glucose, acetate, and car-
bon dioxide as carbon sources.

C 14 -acetate provides a particularly use-
ful measure of the flow in the cycle. With
this tracer it was possible to observe in-

IC^ »• ACETIC

t
ACETIC

+
ACETIC

Fig. 8. Flow of carbon in the various branches
of the Krebs cycle, in micromoles per gram dry
weight of cell per ioo seconds. This flow pat-
tern was observed with E. coli growing exponen-
tially (37 ° C aerated) with glucose acetate and
C0 2 as carbon sources. Other organisms and
other conditions show altered flows. The dashed
lines indicate the dicarboxylic acid cycle, which
is not significant in E. colt.

creased flow rates as the cells started to
grow and came out of the lag phase into
exponential growth. Other carbon sources,
such as glutamic acid, added to the glu-
cose medium caused an immediate altera-
tion of the flow patterns. Still other
changes in the flow patterns were observed
when cells were adapted to use succinic
acid as the sole carbon source.

The flow in the Krebs cycle can be re-
lated to the total utilization of glucose.
Figure 9 shows that only a small fraction
of the glucose carbon enters the Krebs

cycle and that the net C0 2 produced by
the cycle is less than 4 per cent of the total
C0 2 production of the cell. Thus in these
growing organisms, the cycle functions
mainly as a mechanism for amino acid
synthesis and is of only minor importance
in the oxidation of glucose and the produc-
tion of energy.

Recently several investigators in other
laboratories have found that a part of the

420 GLUCOSE-

PYRUVIC ACID

50% C0 2

30% INCORPORATED
20% RELEASED TO MEDIUM,
MAINLY ACETATE

NUCLEIC ACID PENTOSE

SERINE FAMILY

PYRUVIC AMINO ACIDS

MEDIUM

LIPIDS

ASPARTIC FAMILY
GLUTAMIC FAMILY

Fig. 9. When glucose is supplied to exponen-
tially growing E. coli as the sole carbon source,
it is utilized at a rate of 420 micromoles per gram
dry cells per 100 seconds. Approximately 30
per cent of the carbon is incorporated, the re-
mainder being released as C0 2 , acetate, and a
number of other compounds. The synthesis of
the nucleic acid and the various families of amino
acids accounts for a large proportion of the carbon
incorporated. These reactions, however, account
for only a small fraction of the total C0 2 pro-
duced from glucose. The flows indicated are
given in micromoles per gram dry cells per
100 seconds.

glucose is metabolized by a process which
oxidizes the i-carbon. Investigations car-
ried out during the past year with C-i-
labeled glucose show that this oxidation
pathway can contribute only minor quanti-
ties of carbon to the amino acids, as they
contain the full quantity of C-i carbon
which would be expected from the Meyer-
hoff glycolytic system. The oxidation path-

DEPARTMENT OF TERRESTRIAL MAGNETISM

129

way, however, might help to explain the
amount of C0 2 produced, which appears
too large to be accounted for by the other
pathways.

Nucleic Acid Synthesis

The origins of the carbon of the various
constituents of bacterial nucleic acid are
now known, and some features of the
pathways for synthesis of nucleic acid have

PURINE

^ C \ 5

NT C — N

I i C 8
1 s>

CARBON ATOM

2 AND 8

SOURCE
FORMATE

C. C — N

4 AND 5
6

GLYCINE
C0 2

PYRIMIDINE

N /C X *

CARBON ATOM

SOURCE

1 1

C0 2

i'^/'.

4, 5, AND 6

ASPARTIC ACID

Fig. 10. Sources of the carbon atoms of the
purines and pyrimidines in E. coli. The purines
and pyrimidines are organic ring compounds
which, together with a sugar phosphate, make
up the nucleic acids found in all living things.
Although the details of the way in which any
species makes these compounds from its food-
stuffs are not yet known, the usual sources of
their carbon atoms are known. The ring struc-
ture of the compounds and the simple foodstuffs
from which each carbon atom is derived are
shown in this figure.

been established in this laboratory. Two of
the purines, adenine and guanine, con-
tain five carbon atoms. One of these, car-
bon-6, is derived from the carbon dioxide
of the culture medium. Two others, car-
bons 2 and 8, are derived from formate (or
formaldehyde), and the remaining two,
carbons 4 and 5, from the amino acid gly-
cine (fig. 10). Other carbon compounds
may contribute carbon to the bacterial
purines. In some cases the contribution
occurs after the alternative source has been

converted to one ot the primary sources;
in others, the compounds are utilized with
only minor alterations. Thus, serine and
threonine are first converted to glycine,
and the glycine so formed is actually used
in purine synthesis. Adenine serves as a
source for either compound of the bac-
terial nucleic acid; guanine provides only
the required guanine. The pyrimidines
cytosine and uracil contain four carbon
atoms. One of these, carbon-2, is derived
from the carbon dioxide of the culture
medium. Aspartic acid accounts for the
remaining three carbon atoms, carbons 4,
5, and 6 (fig. 10). Since aspartic acid itself
may be synthesized in part from carbon
dioxide, carbon dioxide can supply more
than one carbon of the pyrimidines. Exog-
enous uracil is also utilized in nucleic acid
synthesis. The five-carbon sugar portions,
ribose and desoxyribose, of the nucleic
acids are derived entirely from the glucose
of the culture medium, when cells are
adapted to utilize glucose. When they are
adapted to utilize other carbon sources,
such as succinic or a keto glutaric acids,
the riboses are, of course, derived from
these sources. Even in these cases, how-
ever, the pattern of carbon dioxide incor-
poration into the purines and pyrimidines
remains similar to that described above.

Studies of nucleic acid synthesis with
C 14 -labeled compounds may all be inter-
preted in accord with these findings. For
example, C 14 -acetate contributes radioac-
tivity largely to the pyrimidines of bacterial
nucleic acid. The purines are very weakly
labeled, since acetate cannot readily be con-
verted to C0 2 , formate, or glycine. Analy-
ses of the radioactivity data from a typical
experiment showed that the quantity of
radioactivity found in the pyrimidines after
acetate labeling was exactly equivalent to
the amount expected if the labeling oc-
curred by way of aspartic acid.

130

CARNEGIE INSTITUTION OF WASHINGTON

These studies, together with the results
of isotopic competition studies, provide a
background against which a more com-
plete picture of nucleic acid synthesis may
be drawn. They have also revealed an in-
timate biochemical connection between
proteins and nucleic acid during the
growth of bacteria: glycine and aspartic
acid are precursors of both materials. The
Krebs cycle by which aspartic acid is pro-
duced, therefore, serves as one of the mech-
anisms which link together the synthesis
of protein and nucleic acid.

The Competition Method

Observations of the selective utilization
of various carbon substrates for the syn-
thesis of protein amino acids have led to
the elucidation of the main steps of syn-
thesis for the carbon skeleton of 90 per
cent of amino acid residues of Escherichia
coli protein. The method which has been
used has been termed "isotopic competi-
tion." It may be considered that, in whole
or in part, the carbon chain of the added
substrate molecule competes with carbon
from glucose for a position in the final
protoplasm of the cell. Either the glucose
(always present in the system under study,
as principal carbon and energy source) or
the added substrate may be labeled with
the radioactive isotope C 14 .

The fate of the radioactive C 14 , or by
difference the C 12 , may be determined by
splitting the protein into its constituent
amino acids, and in some cases by degrad-
ing the amino acid molecules, carbon atom
by carbon atom. When the additional sub-
strate (competitor) is labeled with C 14 , a
greater sensitivity for minor carbon path-
ways is obtained. If the competitor is
unavailable with a C 14 label, however, the
experiment may be performed with radio-
active glucose or C 14 2 and an unlabeled
competitor, thus opening up the whole

available supply of organic compounds for
this type of experiment.

The particular value of the competition
method results from the specific response
of the cells to certain substrates. Though
E. coli may be adapted to many substrates
so as to make use of them as sole carbon
and energy sources, there is no evidence
that adaptation plays a role in competition.
The response is due to the changed con-
centration of the chemicals in the environ-
ment rather than to any change in the
enzymatic complement of the cells. Dur-
ing an exposure of one or two generations
of rapid growth a substance will, in gen-
eral, be utilized only in pathways where
it or a very similar compound normally
occurs. Consequently, the competition
method gives information about normal
pathways rather than pathways requiring
adaptation.

Some care has been taken to ascertain
that the presence of competitors has not
so changed the environment that the cell
chemistry operates in wholly new ways.
Cross-check experiments with a wide vari-
ety of tracer compounds and with compet-
itors at various points along a particular
pathway have shown that the biochemical
processes have a great deal of stability. In
only a few cases have major alternative
pathways been established; in other cases,
changes in the rates of flow of carbon have
been observed.

Extracellular Metabolic Products

For the studies of the pathways of amino
acid synthesis it has not been necessary to
have a detailed understanding of the
mechanism of the competition method. In-
vestigations of the mechanism, however,
carried out by observations of the com-
pounds appearing in the growth medium,
have given insight into the cell chemistry,
and have opened up the possibility of

DEPARTMENT OF TERRESTRIAL MAGNETISM ^

studying some aspects of its control and ternally synthesized compound in the pro-
organization, tein.

When E. coli is grown aerobically with In this type of experiment, if unlabeled
glucose as substrate, about 70 per cent of glutamic acid, aspartic acid, glycine, ala-
the glucose carbon appears as extracellular nine, or valine is present as a competitor,
products. The major products are carbon it appears in the medium containing la-
dioxide (50 per cent) and acetate (13 per beled glucose carbon. The amount of the
cent). The remaining small percentage of labeled compound indicates that the rate
the carbon appears in a complex spectrum of synthesis of the compound from glu-
of compounds. About 40 of these have cose has not been strongly reduced. The
been resolved, and several have been iden- presence of unlabeled arginine, leucine, iso-
tified. Normally, none of these compounds leucine, cystine, and lysine as competitors
is predominant, and the amount of any in- causes the respective labeled compounds
dividual compound appearing, such as as- to appear in small quantities, a result
partic or glutamic acid, is very small — less which indicates that the rate of synthesis
than 1 per cent of the amount synthesized is much reduced by the presence of the
by the cells. This finding demonstrates a substance as competitor. On the other
high degree of precision in the control of hand, no trace of radioactivity appears in
the relative rates of synthesis of these com- the medium in the following amino acids
pounds which are used as building blocks when they are present as unlabeled com-
for protein synthesis. Certain variations in petitors : proline, threonine, methionine,
culture conditions may cause a few of these and serine. Thus the rate of synthesis of
compounds to become prominent, indicat- these compounds has been reduced nearly
ing an imbalance in rate of synthesis. to zero by the presence of the competitor.

Most amino acids are utilized by E. coli In some cases precursors of the compet-

for protein synthesis, when they are avail- itor appear in the medium; for instance,

able in the medium. In all these cases the fumaric and succinic acids are produced

incorporation of glucose carbon into the from glucose carbon when aspartic acid is

particular protein amino acid is strongly present as a competitor. This case will be

reduced. In some cases the internally syn- discussed in more detail below. Thus, the

thesized amino acids or compounds which observation of the compounds appearing

occur near by on the synthetic pathways ap- in the medium opens up the possibility of

pear in the growth medium. identifying hitherto unsuspected interme-

It has been possible to observe, and in diates. It also offers the possibility of dem-

most cases identify, the compounds that onstrating that a suspected intermediate is

are produced in the growth medium in re- actually synthesized by the cells. Among a

sponse to the presence of these competitors, series of suspected intermediates, the fol-

If the cells are grown on randomly labeled lowing appear in the medium labeled from

glucose and the competitor is not labeled, glucose when the unlabeled competitor is

substances which derive carbon from the present: succinic, fumaric, and malic acids,

glucose, and, therefore, have actually been and homoserine. Another group of sus-

synthesized by the cells, will be radioactive, pected intermediates which are effective

In this way it has been determined whether competitors gave negative results — that is,

an effective competitor stops the synthesis no trace of radioactive compounds in the

of a compound or simply replaces the in- medium.

132

CARNEGIE INSTITUTION OF WASHINGTON

The observation that succinic, fumaric,
and malic acids appear labeled in the me-
dium when the unlabeled compound is
present is particularly useful, since these
are relatively ineffective as competitors and
the positive demonstration of their synthe-
sis by the cells strengthens the deductions
about the Krebs cycle. The negative re-
sults from the other suspected interme-
diates do not supply decisive evidence on
their role as intermediates, since threonine,
which is synthesized by the cells and is an
intermediate in the synthesis of isoleucine,
also gives a negative result.

Further information about the opera-
tion of the Krebs cycle has been obtained
by comparing two experiments in which
aspartic acid was the competitor. In one
case the glucose was labeled, and in the
other case the glucose was nonradioactive
but labeled C0 2 was present. With the
glucose labeled, radioactive aspartic, fu-
maric, and succinic acids appeared in the
medium. When the C0 2 was labeled,
however, radioactivity appeared only in
the aspartic and fumaric acids. These re-
sults indicate (i) that fumaric acid is the
symmetrical intermediate which is needed
to explain the appearance of C 14 from
C 14 2 in both terminal carbons of aspartic
acid, and (2) that fumaric acid is not
directly converted to succinic acid in these
growing cells.

The excess of fumaric acid is presumably
the cause of the discharge of succinic acid
to the medium, particularly since fumaric
acid added to the medium produces the
same result. This result cannot, however,
be interpreted in terms of the reversal of
the normal reaction succinic — > fumaric,
because very little of the radioactivity of
fumaric acid appears in succinic acid. Ac-
cordingly, it seems that the normal reac-
tion is strongly inhibited, possibly because
the enzyme is saturated with fumaric acid

and no sites are available for adsorption
of succinic acid.

With a relatively small perturbation of
the chemistry of exponentially growing
bacterial cells, considerable insight has been
obtained into a reaction step between two
transient intermediate compounds. The
opportunity to make observations of equi-
librium constants, and possibly other ki-
netic aspects of a particular reaction, in an
only slightly perturbed growing organism
may have considerable biological impor-
tance.

Composition of Escherichia coli

Further information on the composition
of the various fractions which can be ex-
tracted from Escherichia coli has been ac-
cumulated during the year. Observations
of the total uptake of a radioactive tracer
give a crude measure of metabolic activ-
ity. Much more information is obtained if
it is possible to follow the tracer into par-
ticular compounds formed by the cells. In
cases where this is not feasible, it is desir-
able at least to observe the distribution of
the tracer in various broad classes of com-
pounds, which are separated by chemical
fractionation.

The fractionation procedure involves ex-
tracting with cold trichloracetic acid
(TCA), followed by alcohol and by hot
TCA, thereby yielding three soluble frac-
tions and a residual insoluble fraction.

The cold-TCA-soluble fraction contains
most of the transient intermediates of the
cell. Chromatography reveals more than
twenty components (mostly unidentified),
among which glutathione is particularly
prominent. Many of the components re-
act with ninhydrin, but, as they do not
correspond to known amino acids, they are
probably peptides. After hydrolysis, glu-
tamic acid, glycine, cystine, and alanine
appear prominently.

DEPARTMENT OF TERRESTRIAL MAGNETISM

J 33

The alcohol-soluble fraction can be re-
solved into two components, one of which
is lipid or phospholipid and the other ap-
pears to be protein. It seems likely that
these are associated in the cell as a loose
complex which is broken down by chro-
matography. Roughly one-sixth of the cel-
lular protein appears in this fraction. Fur-
thermore, this protein has quite a different
metabolism from the rest of the protein of

TABLE 5
Distribution of radioactivity from various

TRACERS

C14-

glu-

ace-

C> 4 2 S 3 4 P 32 4

cose

tate

Total cell 100 100 100 100 100

Metabolic

intermediates 8 5 4 25 20

Lipids,

phospholipids 15 43 15

Nucleic acids 18 4 33 63

Alcohol-soluble

proteins 10 8 11 12

Residual proteins . . 49 40 52 63 2

the cell, as can be shown by kinetic experi-
ments (see below).

The hot-TCA-soluble fraction (after hy-
drolysis) yields only the constituents of
nucleic acid. The residual insoluble ma-
terial is protein. The distribution of radio-
activity from various tracers among the
fractions is shown in table 5. The amino
acid composition of the protein was deter-
mined by growing cells with C 14 -glucose
as a sole carbon source and measuring the
radioactive content of the separated amino
acids, as shown in table 6.

Utilization of Sulfur Reservoirs for
Protein Synthesis

Escherichia coli utilizes sulfur for pep-
tide and protein synthesis by converting
inorganic sulfur into the amino acids
cysteine and methionine. These bacteria
also utilize other forms of sulfur, includ-
ing the amino acids, for protein synthesis.
Once the sulfur supplied in the medium
has been converted into cell components, it

TABLE 6
Amino acid content of Escherichia coli protein hydrolysates

Amino acid

No.
C

Mol.
wt.

C 14 AA

C 14 val.

Hydrolysate
carbon
- (%)

Mol.
(%)

Percentage

of E. coli

protein

Probable
error
(%)

Alanine

1.34

8.52

13.1

8.72

Arginine

174

1.00

6.36

4.9

7.16

Aspartic acid

133

1.56

9.92

11.5

12.3

Cysteic acid

169

0.14

0.90

1.4

1.9

Glutamic acid

147

2.00
0.49

12.7
3.1

11.8

7.2

14.2
3.8

Glycine

Histidine

155

0.15

0.95

0.7

0.80

Isoleucine

131

1.08

6.9

5.3

5.6

Leucine

131

1.62

10.3

7.9

8.4

Lysine

146

1.16

7.37

5.63

6.60

Methionine

149

0.26

1.65

1.53

1.9

Phenyl-alanine

165

0.60

3.8

2.0

2.7

Proline

115

0.76

4.82

4.46

4.17

Serine

105

0.66

4.2

6.5

5.2

Threonine

119

0.68

4.32

5.0

4.73

Tyrosine

181

0.71

4.7

2.3

3.6

Valine

117

1.00

6.36

5.9

5.45

134

CARNEGIE INSTITUTION OF WASHINGTON

becomes a relatively stable part of the
protoplasm, undergoing exchange only un-
der special conditions and then to a very
limited degree from the TCA-soluble frac-
tion. Glutathione is the chief sulfur con-
stituent of this fraction. Last year it was
reported that the glutathione in the cell
could serve as a source of sulfur for the
proteins. Thus, a stepwise sequence of
synthesis from the sulfur of the medium,
to glutathione, to protein was established.
In the past year it has been found that
during sulfur starvation another chemi-
cally distinct fraction of the cell, the alco-
hol-soluble protein, may also serve as a
source of residual protein sulfur.

Ordinarily, the sulfur composition of
E. coli is a very constant quality. One-
quarter of the sulfur of the cell is contained
in the TCA fraction and three-quarters in
the proteins. The sulfur in the proteins is
contained in the amino acids methionine
and cysteine in the proportion 2 to i. The
alcohol-soluble proteins, which are quali-
tatively different from the residual pro-
teins, constitute about 16 per cent of the
total residual proteins of the cell. The
sulfur content of the bacteria may be
varied, and the distribution of sulfur with-
in the cell altered, by reducing the level of
sulfur supplied in the culture medium, or
by providing chemical forms of sulfur
which do not satisfy all the needs of the
cell. Both conditions result in sulfur star-
vation and eventual cessation of growth.
Immediately after the sulfur supply has
been shut off by one of these means, cells
continue to grow at a normal rate. After
a relatively short period of development,
the growth rate falls with remarkable ab-
ruptness to one-tenth its former value.
Eventually growth stops. During these
periods of growth change, the sulfur con-
tent of the various chemical fractions of
the cell also changes. When the exogenous

sulfur supply is removed and no sulfur is
available, the sulfur in the TCA fraction
decreases sharply, while the protein sulfur
increases. Thus, the sulfur of glutathione
is drawn upon for protein synthesis, acting
as a reservoir. Protein synthesis during
this time occurs at a nearly normal rate.
At the instant when this reservoir has
emptied and the growth rate is altered, the
alcohol-soluble protein undergoes deple-
tion of its sulfur. The alcohol-soluble sul-
fur is then transferred to the residual pro-
tein. The alcohol-soluble fraction also
serves as a reservoir for sulfur, being
drawn upon at a rate proportional to
growth. Experiments with C 14 -labeled al-
cohol-soluble protein show that a flow of
carbon from this material to the residual
proteins is coincident with the flow of sul-
fur. When growth has ceased, neither car-
bon nor sulfur is converted into residual
protein, nor is either of these elements lost
from the cell. The methionine-cysteine
ratio in the residual proteins remains con-
stant all through these activities.

Experiments in which sulfur-deficient
media have been reconstituted at points
along the response sequence have shown
that new protein is made from sulfate
whether sulfate is added during the deple-
tion of the TCA reservoir or after both
reservoirs have been largely emptied. On
the other hand, reconstitution of the me-
dium with methionine after the TCA res-
ervoir is empty has little effect on growth.
Methionine added to the medium at any
time before the instant of depletion of the
TCA-soluble fraction, however, results in
normal growth and the use of methionine
for protein synthesis.

These results add to knowledge of the
metabolism of sulfur by E. coli. They are
of particular significance for the study of
the process of protein synthesis, since they
demonstrate that certain components of the
cell may be sacrificed in times of need to

DEPARTMENT OF TERRESTRIAL MAGNETISM

135

the manufacture of residual protein. One
of these components is a tri-peptide and
another is itself a protein or a very complex
polypeptide. The experiments in which
deficient media were reconstituted indicate
that sulfur-starved cells preserve their pro-
tein-making apparatus and that it will
operate effectively provided it has the ap-
propriate raw materials with which to
work. They also show that even during
starvation protein synthesis remains an or-
derly process: it will not take place with
methionine as the sole available supply of
sulfur; when it does take place, proteins
with methionine and cysteine contents in
the ration 2 to 1 are made. Finally, these
experiments have provided a model system
in which it should be possible critically
to test some fundamental questions about
the mechanism of protein synthesis. For
example, is the alcohol-soluble protein
formed into residual protein by conver-
sions at the macromolecular level, or is this
material degraded to small molecules and
totally reconstructed? Does glutathione
serve merely as a sulfur-atom donor, or
does it function as a sulf ur-peptide source ?
Does the transport of sulfur take place in
the sequence glutathione — alcohol-soluble
protein — residual protein, or do the reser-
voirs function as independent pathways?
These questions provide a challenge to be
faced in the coming months.

Comparative Amino Acid Synthesis

A vast accumulation of chemical analy-
ses of the constituents of a wide variety of
living organisms has led to the realization
of striking biochemical similarities among
all forms of life. This is particularly true
of the proteins, which are found in all
organisms. Furthermore, these compli-
cated molecules are synthesized out of a
common group of amino acid building
blocks.

Does the unity of the biochemical plan

extend to the detailed pathways by which
the amino acids are synthesized? The
theoretical chemist could write many
schemes by which, for instance, isoleucine
might be synthesized. Have organisms
tended to choose a limited number of
specific pathways?

A start has been made in studying this
question. The techniques which have been
evolved for study of synthetic pathways
in E. coli have been successfully employed
using the mold Neurospora crassa, the
yeast Torulopsis atilis, and to a lesser ex-
tent the photosynthetic organism Chlorella
pyrenoidosa.

Neurospora crassa

One of the favorite objects for study of
pathways of amino acid synthesis has been
Neurospora crassa. By use of mutant tech-
niques it has been possible to gather evi-
dence regarding synthetic pathways in this
organism. In addition, the relation be-
tween genetic constitution and synthetic
capacity has been studied. It was particu-
larly important to investigate synthetic
pathways in this organism using a differ-
ent mode of examination, that is, radio-
active tracers. If concordant results were
obtained by two very different methods,
the structure of knowledge regarding the
pathways would be indeed secure. In our
survey of N. crassa some ten different
radioactive tracers have been employed,
namely, C 14 -labeled CO*, CH3COOH, glu-
cose, aspartic acid, glutamic acid, alanine,
glycine, proline, ornithine, and valine.

One of the well established synthetic
pathways in E. coli involves aspartic acid
as a precursor. A similar family has been
found in N. crassa:

Methionine

Aspartate ^ Homoserine

Threonine

a keto butyrate ~^* a keto |3 methyl valerate
~^" Isoleucine

136

CARNEGIE INSTITUTION OF WASHINGTON

The pathways shown are identical with
those found in E. colt. Earlier Neurospora
mutant studies had indicated that methio-
nine and threonine were derived from
homoserine.

A glutamic family similar to that of E.
coli has been found :

Glutamic acid "^Glutamic y semi aldehyde "^Proline

Ornithine ~^" Citrulline ~^ Arginine

Studies with Neurospora mutants had in-
dicated a pathway from glutamic acid to
arginine similar to that shown. The mu-
tant evidence, however, was equivocal con-
cerning the possible role of proline as an
intermediate in arginine synthesis. Studies
with labeled proline indicate that proline is
not ordinarily on the synthetic pathway
of arginine. A small amount of arginine
is derived from proline, however, a fact
which demonstrates that wild-type N.
crassa has the capacity to utilize proline
in arginine synthesis. The existence of
such alternative pathways poses a problem
of interpretation of mutant results and em-
phasizes the desirability of determining
pathways by more than one method.

Synthesis of lysine in N. crassa follows
a pathway differing from that in E. coli.
In the mold, four of the six carbons of
lysine are derived from the Krebs cycle.
Considerations of specific activity show
that the four-carbon moiety involves suc-
cinic acid and that the additional two car-
bons necessary for lysine are obtained from
acetate.

In E. coli it was found that the Krebs
cycle is very important as a source of as-
partic and glutamic acids. Investigations
in this laboratory indicate a similar Krebs
cycle in N. crassa. Earlier work had un-
covered mutants for which such com-
pounds as succinic acid permitted growth.
Our work has supplied considerable addi-
tional evidence. It has been found, in ac-

cordance with predictions of the Krebs
cycle, that (1) C 14 2 is used in synthesis
of aspartic and glutamic acid; (2) the C 14
of such glutamic acid is located in the
carboxyl carbon; (3) in the cycle, carboxyl-
labeled glutamic acid is decarboxylated;
(4) CH 3 C 14 OOH is utilized in synthesis
of glutamic acid and to a lesser extent in
synthesis of aspartic acid; (5) aspartic acid
can be converted into glutamic acid; (6)
glutamic acid can be converted into as-
partic acid.

Another pathway of synthesis found in
E. coli is as follows:

Pyruvic acid

Alanine
^ Valine ^ Leucine

A similar sequence has been found in N.
crassa. A sequence involving serine is pres-
ent in both organisms:

^^-^T Glycine
Serine -^^

^^^ Cysteine

N. crassa and E. coli differ in their be-
havior toward exogenous threonine. Both
organisms use this substance in the syn-
thesis of isoleucine. In addition, E. coli can
degrade exogenous threonine to glycine,
whereas N. crassa does not.

Torulopsis utilis

The yeast Torulopsis utilis is another
organism which has been a favorite object
of study. In particular, a number of in-
vestigators have studied metabolism of iso-
topic acetate and C0 2 . Work in this lab-
oratory has served to confirm and extend
this earlier work. In addition to acetate
and CO2, some ten other labeled sub-
stances have been used.

Aspartic and glutamic families closely
similar to those found in N. crassa have
been demonstrated. The synthesis of argi-
nine parallels that in E. coli, and our re-
sults confirm and extend earlier observa-

DEPARTMENT OF TERRESTRIAL MAGNETISM

137

tions. Studies of lysine synthesis indicate
that the pathway is similar to that in N.
crassa and confirm earlier observations on
T. utilis.

Further evidence was found for existence
of a Krebs cycle as in E. coli and N. crassa.
The observations were of the same type
as the six cited in the discussion of N.
crassa.

The close similarity of T. utilis and N.
crassa extends to the two synthetic se-
quences involving pyruvic acid and serine
as key precursors.

The pathways by which the sulfur amino
acids are synthesized are not identical in
the bacterium, yeast, and mold, though
common features do exist. In all cases the
carbon skeleton of cysteine is derived from
serine, and four of the carbons of methio-
nine originate from aspartic acid via homo-
serine. Differences are found in the ease of
utilization of inorganic sulfur sources and
particularly in the metabolism of organic
sulfur compounds. For instance, N. crassa
and T. utilis can use methionine as a sole
sulfur source, whereas E. coli cultures do
not grow under such circumstances. Ex-
ogenous cystathionine is used by the mold
and yeast but not by E. coli.

Information regarding cellular organi-
zation is a by-product of isotopic competi-
tion studies. Investigations of the aspartic
family revealed a similar degree of compe-
tition by aspartic acid, homoserine, threo-
nine, a keto butyric acid, a keto (3 methyl
valeric acid, and isoleucine in the three
respective organisms. On the other hand,
exogenous glutamic acid, which is an effec-
tive competitor in E. coli, has little effect
in N. crassa or T. utilis. Other members
of the glutamic family compete more suc-
cessfully in E. coli than in mold or yeast,
although in the case of proline the differ-
ence is less marked. It appears that in
almost all respects the biosynthetic path-
ways and organization of N. crassa and T.

utilis are identical. In many ways this pair
is similar to E. coli, though differing in a
few respects.

Studies of the pathways of amino acid
synthesis in Chlorella pyrenoidosa are
rather fragmentary. The pattern of acetate
incorporation is strikingly similar to that
found in N. crassa. Dark fixation of C 14 2
gives rise to labeled aspartic and glutamic
acids, thus making it likely that a Krebs
cycle is operative in this organism.

These investigations of the comparative
biosynthetic pathways point anew to the
existence of great similarities in the bio-
chemical plan of living things. At the
same time, it is also clear that differences
in synthetic patterns exist and that some
caution must be employed in transferring
results from one organism to another.

Biological Processes in Escherichia coli

The biological syntheses observed by the
use of isotopes, isotopic competition, and
chromatography have been given quanti-
tative expression in terms of the flow of
carbon, sulfur, and phosphorus. These
syntheses begin with relatively simple raw
materials and yield finally an extraordi-
nary variety of organic chemicals. Yet liv-
ing bacteria are more than a bewildering
collection of complex chemicals derived
from sugar, salts, and water. They are re-
sponsive mechanisms, reacting instantly
and sometimes violently to subtle changes
in their environment. They tend to resist
change rather than undergo change, but
do adjust when the stresses are great, and
successfully preserve integrated activities
in differing environments. Although these
are qualities not always interpretable in
exact terms, they are among the most ob-
vious characteristics of living things. Many
of their facets have been examined in this
and earlier reports, and they have become
perhaps less mysterious as a result.

138

CARNEGIE INSTITUTION OF WASHINGTON

The bacteria which have been studied
are freely permeable to most metabolites.
Consequently, the protoplasm of the cell
is in intimate contact with the foods in its
environment and may be expected to re-
spond biochemically to chemical changes
in the nutritive medium.

When a growing cell is presented with
a metabolite which is particularly suitable
for synthesis, the cell utilizes it, ordinarily
without delay, to make the chemicals it
needs. In most cases where the metabolite
is one that the cell would have made and
used if left to its own devices, no adjust-
ment in its biochemical apparatus is found.
The cell has not distinguished between ex-
ogenous and endogenous chemicals. Al-
though this behavior is not necessarily to
be expected in such a highly integrated
chemical system, it is the usual situation,
and one which has made isotopic competi-
tion an effective tool. In exploiting the in-
ability of the cell to make this distinction,
it has been found possible to introduce
chemicals one at a time into the operations
of the cell and by following their conver-
sion learn about normal mechanisms and
processes. The biological syntheses per-
formed by the cell generally result in the
formation of new protoplasm, thereby
"bottling up" energy rather than releasing
it through degradative processes. The
Krebs cycle serves as a striking example of
this property. The conversions which take
place in this cycle and in many other cel-
lular processes are highly specific and are
related in precise ways to the chemical na-
ture of the compound fed and the com-
pounds formed. The specificity of the
conversions which occur contrasts with the
situation in other organisms, especially
adult forms of higher animals, where rela-
tively nonspecific utilization of metab-
olites is the rule. In the bacterium the
economy of the cell appears largely to be
devoted to synthesis, rather than to replace-

ment or repair. Thus, there is little ex-
change of the constituents of protein or
nucleic acid with components in the cul-
ture medium, and what has been called
the "dynamic state" has relatively little
meaning in the biochemistry of growing
E. coli.

Some metabolites supplied in the culture
medium are distinguished from similar
materials made and used by the cell. In
these cases the cell changes its biochemical
activities in immediate response to the
presence of the added material. For ex-
ample, E. coli makes threonine and uses it
for isoleucine synthesis as well as to con-
struct protein. When the bacteria are sup-
plied threonine in ordinary culture media,
they carry out these functions with it and
in addition convert it to glycine. The gly-
cine so formed is used in peptide, protein,
and nucleic acid synthesis. The bacteria
respond to the presence of threonine by
calling into action a "glycine-from-threo-
nine" process. This process is not nor-
mally operative, yet the capacity to bring
about the conversion is present. How the
cell keeps this capacity from being ex-
hibited in the absence of exogenous threo-
nine is not known. A reasonable hypothe-
sis, however, is that bacterial threonine is
bound into an active complex which is not
convertible to glycine. The experiments on
threonine, while helping to explore some
amino acid pathways, have thus also
yielded some insight into the chemical or-
ganization of the cell. When such insight
can be formulated in terms of spatial ar-
rangements of parts and sequences of
events, we may hope to understand how a
living system integrates the diverse proc-
esses which do occur, and keeps other pos-
sible activities from occurring.

It is a common observation that the
kinds and amounts of chemicals which
comprise the bacterial cell remain remark-
ably constant in spite of radical changes in

DEPARTMENT OF TERRESTRIAL MAGNETISM

139

the environment in which the cell develops.
Alterations in the constituents of the cell
are in general progressively less as struc-
tures become more highly organized.

Chemically, the bacterial cell may be
visualized as a series of classes of chem-
icals, the members of the classes becoming
increasingly larger and more complex
along the series. Thus, in the water space
within the cell are all the constituents pres-
ent in the medium. They include water,
salt, sugar, and a host of other organic
compounds of relatively low molecular
weight. In another class are the chemicals
which do not permeate freely but which
may be extracted from the cells by means
of cold dilute acid. These include pep-
tides, such as glutathionine, and other
complex organic molecules. These com-
pounds are part of the organized proto-
plasm of the cell, but their associations
with proteins or nucleic acid are easily
disrupted. In still another class, the alco-
hol-soluble chemicals, are found lipids and
complex polypeptides. The polypeptide
material contains the same amino acids in
nearly the same proportions as the rest of
the bacterial protein. Finally, there are the
truly macromolecular components, the re-
sidual proteins and the nucleic acids.

The water-space components reflect
every chemical change in the environment.
Thus, if sodium salts in the medium are
replaced by potassium analogues, potas-
sium salts are found in the cell. Changes
in the biochemical activity of the cell are
also reflected by the water-space constitu-
ents, and through such changes the cell
alters the environment in which it devel-

ops. For example, a medium containing
glucose and homoserine supplied to grow-
ing cells will be found to contain appre-
ciable quantities of organic compounds de-
rived from glucose, and specifically related
to the metabolism of homoserine and to
the synthesis of methionine and isoleucine.
These compounds are not found in the
absence of homoserine.

The acid-soluble portion is somewhat
less variable in relation to changes in the
environment. Nevertheless, it will change
whenever the stresses are large enough.
When E. coli is starved for sulfur by re-
moval of the supply in the medium, the
sulfur of glutathione is utilized for the
synthesis of protein methionine and cys-
tine, whereas normally it does not function
in this way. On continued sulfur starva-
tion, even the alcohol-soluble materials are
attacked and utilized for making residual
protein. Finally, all protein synthesis
ceases and growth stops. In spite of these
extreme conditions, the residual proteins
and nucleic acids have not undergone
qualitative change, and their relative pro-
portions remain constant.

Processes which give rise to variation,
as well as mechanisms which prevent
change, are to be found at every level of
the biochemical activities of the bacterial
cell. They may be correlated with the de-
gree of complexity of the collections of
chemicals considered, and depend for de-
tail on the particular activity studied.
Whatever their specific qualities, they
are orderly, integrated processes. They
may be controlled, altered, and sometimes
predicted.

OPERATIONS AND STAFF

Co-operative Work of the Department of the National Research Council, Na-

The Department has pursued the Insti- tional Institutes of Health, National Bu-

tution's policy of co-operation by engaging reau of Standards, Department of De-

in research activities with representatives fense, Geological Survey, Geophysical In-

140

CARNEGIE INSTITUTION OF WASHINGTON

stitute of Huancayo (Peru), Catholic and
Cornell Universities, University of Alaska,
Univeisity of Virginia, and commercial
organizations. Close liaison has been main-
tained with international scientific organi-
zations such as the International Union
of Geodesy and Geophysics and the Inter-
national Scientific Radio Union. We have
had visiting investigators, some on fellow-
ship appointments, from various countries
including Denmark, England, Holland,
Norway, Scotland, and South Africa.

Our joint investigations of seismic phe-
nomena and the age of igneous intrusives
have been continued with the Geophysical
Laboratory. We are indebted to the Na-
tional Bureau of Standards and the De-
partment of the Navy and to many indi-
vidual collaborators in this country and
abroad for assistance in the radio astron-
omy programs. The Department of the
Navy has also continued to help with our
seismic program.

Assistance is being given to the Uni-
versity of Western Australia in a project
for using isotope and mass-spectrograph
procedures for the determination of ages of
rocks in Australia.

Cosmic-ray meters have been continued
in operation at widely distributed points
through the generous co-operation of the
observatories at Huancayo, Peru; Christ-
church, New Zealand; Godhavn, Green-
land; and Cheltenham, Maryland.

Government contracts, as mentioned in

previous reports, have been continued,
without subsidy, for investigations of the
earth's crust, cosmic rays, and the meas-
urement of the ages of igneous intrusives
and other minerals, especially in Pre-
Cambrian rocks. One staff member did
research work for the government for the
entire report year, and one to July 15, 1952,
both on full-time leave of absence; other
staff members have acted as consultants for
short periods.

Administration and Operation

We have continued to edit the Journal
of Geophysical Research, and the Institu-
tion has continued to meet part of the pub-
lication expenses.

Lectures

G. M. Temmer gave a series of twelve
informal lectures for the staff during De-
cember and January on "Physics of the
nucleus in the medium and low energy
region."

Lectures were given by visitors as fol-
lows:

September 18, 1952, "Physics at the Uni-
versity of Zurich," Hans H. Staub.

October 6, 1952, "Terrestrial magnet-
ism," S. K. Runcorn.

November 20, 1952, "Two-wavelength
microscopy," M. J. Buerger.

January 21, 1953, "The chemical com-
position of the primary cosmic radiation,"
D. M. Ritson.

BIBLIOGRAPHY

Abelson, P. H. Amino acid sequence in pro-
teins. (Abstract) Science, vol. 115, p. 479
(1952).

E. T. Bolton, and E. Aldous. Utiliza-
tion of carbon dioxide in the synthesis of
proteins by Escherichia coli. I. Jour. Biol.
Chem., vol. 198, pp. 165-172 (1952).

Utilization of carbon di-

oxide in the synthesis of proteins by Escher-
ichia coli. II. Jour. Biol. Chem., vol. 198,

pp. 173-178 (1952).
— See Bolton, E. T.; Roberts, R. B.

Ahrens, L. H. See Aldrich, L. T.; Herzog, L. F.
Aldous, E. See Abelson, P. H.; Bolton, E. T.
Aldrich, L. T., J. B. Doak, and G. L. Davis.
Mineral age measurement: Mass spectromet-

DEPARTMENT OF TERRESTRIAL MAGNETISM

I 4 I

ric determinations of Rb 87 and Sr 87 in le-
pidolites. (Abstract) Bull. Geol. Soc. Amer.,
vol. 63, p. 1230 (1952).

The use of ion exchange

columns in mineral analysis for age determi-
nation. Amer. Jour. Sci., vol. 251, pp. 377-

387 (i953)-

— L. F. Herzog, J. B. Doak, and G. L.
Davis. Variations in strontium isotope abun-
dances in minerals. Part 1: Mass spectro-
metry: analysis of mineral sources of stron-
tium. Trans. Amer. Geophys. Union, vol.
34, pp. 457-460 (1953).

W. K. Holyk, F. B. Whiting,

and L. H. Ahrens. Variations in isotopic
abundances of strontium. Phys. Rev., vol.
89, pp. 631-632 (1953).

— See Davis, G. L.; Herzog, L. F.; Nico-

ion-exchange columns. (Abstract) Bull.
Geol. Soc. Amer., vol. 63, p. 1242 (1952).
— See Aldrich, L. T.; Keith, M. L.

LAYSEN, L. O.

Bolton, E. T., and P. H. Abelson. Utilization of
carbon dioxide in the synthesis of protein
and nucleic acid by Escherichia coli. (Ab-
stract) Science, vol. 115, p. 479 (1953).

and E. Aldous. Utilization of

carbon dioxide in the synthesis of nucleic
acid by Escherichia coli. Jour. Biol. Chem.,
vol. 198, pp. 179-183 (1952).

R. Britten, and D. B. Cowie. Synthesis

of the aspartic and glutamic families of
amino acids by Escherichia coli. (Abstract)
Science, vol. 117, p. 465 (1953).

See Abelson, P. H.; Cowie, D. B.; Rob-

erts, R. B.
Britten, R. The scattering of 31.5-Mev protons
from several elements. Phys. Rev., vol. 88,
pp. 283-294 (1952).

See Bolton, E. T.

Brown, H. S. See Patterson, C.

Cowie, D. B., and E. T. Bolton. Sulfur metab-
olism in Escherichia coli. IV. Influence of
nonmethionine sulfur compounds upon a
methionine requiring mutant. Jour. Bac-
teriol., vol. 64, pp. 87-96 (1952).

N. P. Heydenburg, and G. C. Phillips.

Inelastic scattering of protons from light
nuclei. Phys. Rev., vol. 87, pp. 304-306
(1952).

See Bolton, E. T.

Davis, G. L., and L. T. Aldrich. Determination
of the age of lepidolites by the method of
isotope dilution. Bull. Geol. Soc. Amer.,
vol. 64, pp. 379-380 (1953).

Mineral age measurement: Chem-
ical separations of Rb and Sr by means of

Doak, J. B. See Aldrich, L. T.; Keith, M. L.;
Nicolaysen, L. O.

Ferraro, V. C. A. On the theory of the first
phase of a geomagnetic storm: A new il-
lustrative calculation based on an idealised
(plane not cylindrical) model field distribu-
tion. Jour. Geophys. Res., vol. 57, pp. 15-

49 (1952).

Forbush, S. E. See Neher, H. V.

Graham, J. W. Changes of ferromagnetic min-
erals and their bearing on magnetic proper-
ties of rocks. Jour. Geophys. Res., vol. 58,
pp. 243-260 (1953).

(Discussion). Rev. Mod. Phys., vol. 25,

pp. 295-296 (1953).

Exsolution phenomena and the magnetic

properties of rocks. (Abstract) Science, vol.
117, p. 3044 (1953).

Note on the significance of inverse mag-

netizations of rocks. (Letter to editor) Jour.
Geophys. Res., vol. 57, pp. 429-431 (1952).
Herzog, L. F., L. T. Aldrich, W. K. Holyk,
F. B. Whiting, and L. H. Ahrens. Varia-
tions in strontium isotope abundances in
minerals. Part 2: Radiogenic Sr 87 in biotite,
feldspar, and celestite. Trans. Amer. Geo-
phys. Union, vol. 34, pp. 461-470 (1953).

See Aldrich, L. T.

Heydenburg, N. P., and G. M. Temmer. Alpha-
particle scattering in nitrogen. Phys. Rev.,
vol. 91, p. 439 (1953).

See Cowie, D. B.; Rose, W. B.; Talbott,

F. L.; Temmer, G. M.

Holyk, W. K. See Aldrich, L. T.; Herzog, L. F.

Hudspeth, E. L. See Rose, W. B.

Inghram, M. G. See Patterson, C; Tilton,

G. R.

Keith, M. L., M. A. Tuve, G. L. Davis, and J. B.
Doak. Ratio of oxygen isotopes in quartz
of contrasted origin. (Abstract) Bull. Geol.
Soc. Amer., vol. 63, p. 1270 (1952).

Larsen, E. S. See Patterson, C.

Neher, H. V., and S. E. Forbush. Correlation of
cosmic-ray ionization measurements at high
altitudes, at sea level, and neutron intensi-
ties at mountain tops. Phys. Rev., vol. 87,
pp. 889-890 (1952).

Nicolaysen, L. O., L. T. Aldrich, and J. B.
Doak. Age measurements on African micas
by the strontium-rubidium method. (Ab-
stract) Trans. Amer. Geophys. Union, vol.
34, p. 342 (1953).

142

CARNEGIE INSTITUTION OF WASHINGTON

Parkinson, W. C. Note on the concentration of
condensation nuclei over the western At-
lantic. (Letter to editor) Jour. Geophys.
Res., vol. 57, pp. 314-315 (*95 2 )-

Patterson, C., H. S. Brown, M. G. Inghram,
E. S. Larsen, and G. R. Tilton. The dis-
tribution of lead and uranium in a precam-
brian granite. (Abstract) Trans. Amer. Geo-
phys. Union, vol. 34, p. 343 (1953).

See Tilton, G. R.

Phillips, G. C. See Cowie, D. B.

Roberts, R. B., and P. H. Abelson. The role of
the tricarboxylic acid cycle in amino acid
synthesis in Escherichia coli. (Abstract) Sci-
ence, vol. 117, p. 471 (1953).

and E. T. Bolton. The role of gluta-
thione in protein synthesis by Escherichia
coli. (Abstract) Science, vol. 115, p. 479
(1952).

Rose, W. B., E. L. Hudspeth, and N. P. Heyden-
burg. Neutrons from deuteron bombard-
ment of N 14 . Phys. Rev., vol. 87, pp. 382-
383 (1952).

Scott, W. E. List of recent publications. Jour.
Geophys. Res., vol. 57, pp. 318-322 (1952);
vol. 57, pp. 435-438 (1952); vol. 57, pp.
543-546 (1952); vol. 58, pp. 121-126 (1953);
vol. 58, pp. 291-294 (1953).

Talbott, F. L., and N. P. Heydenburg. Gamma-
ray resonances in the alpha-particle bom-
bardment of beryllium and boron. Phys.
Rev., vol. 90, pp. 186-187 (1953).

Temmer, G. M., and N. P. Heydenburg. Preci-
sion alpha-alpha scattering at low energies.
II. Phys. Rev., vol. 90, p. 340 (1953).

See Heydenburg, N. P.

Tilton, G. R., C. Patterson, and M. G. In-
ghram. Mass spectrometric determination of
thorium. (Abstract) Bull. Geol. Soc. Amer.,
vol. 63, p. 1305 (1952).

See Patterson, C.

Tuve, M. A. See Keith, M. L.

Vestine, E. H. On variations of the geomag-
netic field, fluid motions, and the rate of
the earth's rotation. Proc. Nat. Acad. Sci.,
vol. 38, pp. 1030-1038 (1952).

On variations of the geomagnetic field,

fluid motions, and the rate of the earth's
rotation. Jour. Geophys. Res., vol. 58, pp.
127-145 (1953).

Wells, H. W. F-region effects of solar eclipse
at sunrise, Sept. 1, 1951. (Abstract) Trans.
I. R. E., Prof. Group on Antennas and
Propagation, No. 3, p. 210 (1952).

Ionospheric effects of solar eclipse at sun-
rise, Sept. 1, 195 1. Jour. Geophys. Res., vol.
57, pp. 291-304 (i95 2 )-

Radio interferometer technique applied

to measurement of meteor velocities. (Let-
ter to editor) Jour. Geophys. Res., vol. 58,
pp. 284-286 (1953).
Whiting, F. B. See Aldrich, L. T.; Herzog,
L. F.

STAFF AND ORGANIZATION

Scientific Staff

Director: M. A. Tuve.
Staff Members

Geophysics: L. T. Aldrich, S. E. Forbush,
J. W. Graham, E. A. Johnson,* H. E. Tatel,
G. R. Tilton, E. H. Vestine, H. W. Wells.
Laboratory and Biophysics: P. H. Abelson,
E. T. Bolton, R. J. Britten, D. B. Cowie,
N. P. Heydenburg, R. B. Roberts, G. M.
Temmer.
Guests, Associates, Fellows, and Visiting Investi-
gators: S. Berko, University of Virginia; H. G.
Booker, Cornell University; J. G. Brennan,
Catholic University; Miss B. W. Catlin, Mar-
quette University School of Medicine; H. R.
Crane, University of Michigan; Mrs. Josephine
K. Doherty, National Institutes of Health;
W. R. Duryee, National Institutes of Health;

* On leave of absence for government work.

C. T. Elvey, University of Alaska; G. Field,
Princeton University; J. W. Findlay, Cavendish
Laboratory, Cambridge, England; Miss J. Gil-
lespie, National Institutes of Health; P. J.
Hart, Harvard University; F. Hereford, Uni-
versity of Virginia; H. W. Koch, National
Bureau of Standards; B. Kok, Agricultural Col-
lege, Holland; C. Levinthal, University of
Michigan; Soren Lovtrup, Carlsberg Labora-
tory, Copenhagen, Denmark; C. J. Marhoefer,
Catholic University; Kenneth McQuillen, Uni-
versity of Cambridge, Cambridge, England;
Leif Owren, Institute of Theoretical Astro-
physics, Oslo, Norway; Leon T. Parker, De-
partment of the Navy; A. T. Price, Royal
Technical College, Glasgow, Scotland; J. A.
Ratcliffe, Cavendish Laboratory, Cambridge,
England; D. M. Ritson, University of Roch-
ester; Mrs. I. Z. Roberts; S. K. Runcorn, Uni-
versity of Cambridge, Cambridge, England;

DEPARTMENT OF TERRESTRIAL MAGNETISM

143

Martin Ryle, Cavendish Laboratory, Cam-
bridge, England; F. G. Smith, Cavendish Lab-
oratory, Cambridge, England; H. H. Staub,
University of Zurich, Switzerland; M. Sugiura,
University of Alaska; F. L. Talbott, Catholic
University; G. M. Temmer, National Bureau
of Standards; R. E. Trumble, Jr., Catholic
University; H. C. van de Hulst, Leiden Ob-
servatory, Holland; H. J. Vogel, Yale Uni-
versity.

Operating Staff

Administrative: M. B. Smith, W. F. Steiner.

Office and Clerical: Miss I. Debber,t W. N.
Dove, Miss E. F. French, W. C. Hendrix,
D. J. O'Rourke, Miss H. E. Russell, Mrs. F. B.
Silberstein, L. Tupler,J Mrs. E. A. Walck.t

X Resigned.

Instrument Shop: B. J. Haase, L. A. Horton,
J. G. Lorz.

Research Assistants and Laboratory Assistants:
S. J. Buynitzky, H. E. Cronin, J. B. Doak,
E. T. Ecklund, R. E. Hewitt, P. A. Johnson,
C. J. Ksanda, Mrs. J. G. Lappin, C. A.
Little, Jr., Mrs. E. A. MacKenzie, M. H.
MacKenzie, P. F. Michelsen,t Miss B. K.
Phillips, R. W. Reuschlein, W. E. Scott.

Computer: Miss I. Lange.

Maintenance: C. Balsam, C. R. Domton, C. P.
Giffin, M. O. Pedersen, E. Quade, S. Swant-
kowski.

Part-time and Temporary Employees: Seventeen
part-time and temporary employees were en-
gaged during the year, usually for short pe-
riods, to assist in the office and laboratory
work.

DEPARTMENT OF PLANT BIOLOGY

St an j or d, California
C. STACY FRENCH, Director

It has long been known that chlorophyll
is the principal absorber of sunlight as it
falls on the green leaves of plants. There
are also present in all plants other pig-
ments, such as carotenoids and the phyco-
bilins of various algae, which play a part
in the photosynthetic process. It appears
highly probable, however, that the function
of these accessory pigments is merely that
of light absorbers and that all they do with
the energy which they take up is to pass
it along to chlorophyll; that is, that they
act solely by supplementing the light-ab-
sorbing role of chlorophyll. A chlorophyll
molecule, after absorbing light or receiving
it by transfer from other pigments, has
more energy than before, and in some way,
probably by means of enzyme systems,
this energy is handed on to water mole-
cules. The result is a splitting of water to
free oxygen and active hydrogen that can
be used, through enzymatic chains, to re-
duce carbon dioxide. This is the vital
process by which green plants convert the
energy of sunlight to usable chemical
energy. Just how chlorophyll and the un-
known components of this system work to-
gether in this remarkable chain of events
is only partly understood.

Several of the great schools of organic
chemistry have studied the chemical struc-
ture of chlorophyll with the hope of elu-
cidating its function in photosynthesis.
This structure has been established, but
the means by which chlorophyll acts re-
mains unknown. Another approach to the
problem is the study of the photochemical
reactions of chlorophyll when dissolved in
organic solvents. Such studies are actively
progressing in a number of laboratories

and have resulted in much information
about the chemical changes caused in other
substances by light absorbed by chloro-
phyll. The changes that chlorophyll itself
undergoes during these transformations are
of particular interest in relation to the
problem of the mechanism by which it
acts in photosynthesis. Reversible changes
in its absorption spectrum caused by light
absorption, or by changes in its oxidation
state, show that it does more than merely
absorb light and pass on the energy.
Chlorophyll itself, at least in reactions car-
ried out in organic solvents, actually goes
through reversible chemical changes much
as do the coenzymes that carry on other
metabolic reactions in living cells.

Extracted chlorophyll in organic solvents
behaves very differently from chlorophyll
in its natural state in chloroplasts. A great
deal of work has gone into the detailed
study of this difference without as yet mak-
ing entirely clear the chemical nature of
the chlorophyll complex in the form in
which it does its work. It has long been
hoped that reversible changes in the ab-
sorption spectrum of chlorophyll could be
found in live plants, thus making it pos-
sible to measure directly some chemical
change in native chlorophyll as it carries
out photosynthesis and as it is influenced
by environmental conditions.

Such an effect has been discovered dur-
ing the past year, not yet in green plants,
however, but in the purple bacteria, which
have a somewhat different kind of chloro-
phyll. The decrease of the main absorption
peak and the shift of a minor peak of bac-
teriochlorophyll in living bacteria when
illuminated were found by Dr. L. N. M.

146

CARNEGIE INSTITUTION OF WASHINGTON

Duysens, of the Biophysical Research
Group at the University of Utrecht. Dr.
Duysens is continuing the study of this
phenomenon as a Research Fellow in this
Department. He has recently found that
the change in absorption is much greater
when oxygen is added to the cell suspen-
sion. The effect is also observed in water
suspensions of disintegrated bacteria in the
presence of an oxidant, and the effect is
abolished by addition of a reductant. Ap-
paratus has been devised for the study of
the time course of this change, which ap-
pears within a second after illumination
and disappears within a second after the
light is off. The further exploration of
this effect should lead to a much clearer
understanding of the means by which
chlorophyll participates in photosynthesis.
Dr. Smith's studies on the rapid forma-
tion of chlorophyll from protochlorophyll
when dark-grown plants are exposed to
light have clarified a previously puzzling
situation. Some years ago the action spec-
trum for the transformation of proto-
chlorophyll to chlorophyll showed that
protochlorophyll has its absorption peak at
650 mu in live etiolated corn leaves. It
was found, however, that a maximum in
the fluorescence spectrum of protochloro-
phyll in etiolated leaves came at about
638 mjj rather than at slightly over 650, as
would be expected. Recent experiments
have shown a slow change from a 650- to
a 635-mM absorption peak in glycerine ex-
tracts of etiolated leaves. This change takes
place at room temperature, but not at o° C,
in a few hours. Since the fluorescence
measurements had necessarily been made
with leaves inactivated by heating, the ap-
parent discrepancy is now believed to have
been due to the change from the native
form of protochlorophyll to the altered
form having the 635-mM band. In collabo-
ration with Dr. Robert Hill it was found
that some seed coats of squash showed

fluorescence spectra of only the normal
form of protochlorophyll, whereas others
had both the normal and the altered form.
It is not yet certain whether the altered
form can be transformed to chlorophyll
when illuminated. The altered form may
simply be protochlorophyll that has been
split off from its normal protein carrier
so that it no longer exists as part of the
holochrome complex. The time course of
the protochlorophyll-to-chlorophyll trans-
formation follows with high precision the
rate equation for a bimolecular reaction
when both reactants are present in equal
concentrations, but there is so far no very
clear understanding of the reason for this
situation. Measurements of the rate of pro-
tochlorophyll transformation have been
made over the whole range from —195°
to + 55 ° C. The fact that the process takes
place reasonably rapidly far below the
freezing point of water shows that it must
go through very few if any enzyme-cata-
lyzed steps. The optimum temperature for
this transformation lies near 39° C. Above
this temperature the reaction is heat-inac-
tivated according to a pattern resembling
the heat denaturation of proteins. Water
suspensions of disintegrated bean leaves
will still transform their protochlorophyll
to chlorophyll, but barley leaves must be
disintegrated in glycerine or in concen-
trated sucrose solutions rather than in
water to retain this activity.

Light has many effects on plants in addi-
tion to supplying the driving power for
photosynthesis; it is used for chlorophyll
formation, for determining the orientation
of the parts of plants above ground, and
for controlling the size and shape of the
various parts of plants, besides influencing
the ability of plants to produce flowers
and of seeds to germinate. Another effect
of light is to cause a change in the con-
sistency of the material inside plant cells.
Different colors of light have very diverse

DEPARTMENT OF PLANT BIOLOGY

147

effects on these different reactions, which
means that the absorbing pigments in-
volved in the reactions are not all the same.
The very small concentration of some of
the very active pigments in the plant cells
makes it difficult and often impossible to
isolate and identify them. The quantitative
effect of any particular color of light is,
however, except for internal screening
effects of other pigments, proportional to
the absorbing capacity of the pigment for
that particular wave length. Thus by meas-
uring the response produced by many dif-
ferent wave lengths it is possible to deduce
the absorption spectrum and hence at least
partially to identify the active pigments
causing the various reactions of plants to
light. Experiments in progress by Dr.
Hemming I. Virgin show that only blue
and blue-green light cause the change in
the viscosity of cytoplasm discovered some
years ago in water plants and recently
found in land plants as well. There are
difficulties in the precise measurement of
this effect that have so far made the exact
shape of the action spectrum hard to deter-
mine, but the results obtained show that
the pigment responsible is very likely a
carotene or a riboflavin type of compound
such as is also found to be active in the
phototropic orientation of plants. The
mechanism of the change of consistency
has been shown to involve three separate
steps: a photochemical reaction, followed
by a slower reaction independent of light,
and a still slower back reaction which
diminishes the effect over a period of sev-
eral minutes. The latter reaction is slowed
by lowering the temperature. Though this
change in viscosity seems to be of general
occurrence, its physiological value, if any,
is not known. It is certainly reasonable to
suppose that the large changes of proto-
plasmic viscosity induced by light in
plant cells could drastically alter the rate

and perhaps the nature of their metabolic
processes.

Much has been written about the pos-
sible use of Chlorella as human food, but
very little Chlorella has actually been eaten.
Some experiments on the addition of
Chlorella ellipsoidea, produced in Japan,
to soups, breads, jelly rolls, noodles, and
ice cream were made by Professor and
Mrs. Hiroshi Tamiya. The foods were
prepared in batches containing different
amounts of Chlorella and as controls with-
out any. Generally the foods were im-
proved in taste by Chlorella, and particu-
larly the enriched breads and ice cream
were highly palatable to Japanese, Ameri-
can, and European subjects. Thus the
direct addition of Chlorella to food seems
feasible and could add significant amounts
of protein without unpleasant tastes.

A study of the growth rate of Chlorella
in intermittent light such as might be ob-
tained in turbulent cultures has been com-
pleted this year by Professor Jack Myers
and Dr. J. Neal Phillips at the University
of Texas. The program at Texas for the
selection from nature of algal strains par-
ticularly suitable for use in mass culture
has been intensified and expanded to in-
clude a search for nitrogen-fixing blue-
green algae as well as for Chlorella strains
having a high temperature tolerance and a
high efficiency in bright light. One of the
high-temperature strains is being tried in
Professor Tamiya's pilot plant in Tokyo.

The range-grass program carried on in
co-operation with the U. S. Soil Conserva-
tion Service has reached the stage where
the new methods of interspecific hybrid
formation and the stabilization of non-
sexual new strains have been thoroughly
established as a useful addition to the
standard methods of plant breeders. This
program has resulted in the production of
numerous hybrids that give promise of
having practical value as cattle forage, par-

148

CARNEGIE INSTITUTION OF WASHINGTON

ticularly under the western range condi-
tions for which they were selected. During
the past year tests of the thirty selected
strains previously reported have been con-
tinued and expanded. In addition, some
new combinations of the selected hybrids
have been found to show characteristics
that greatly encourage further efforts to
test adequately their comparative value as
forage crops.

The planning, execution, and evaluation
of the large-scale field tests essential for in-
troducing the new grasses into agricultural
use is clearly a project of great size, requir-
ing specialized facilities. We are planning
arrangements for further collaboration in
testing as thoroughly as possible various
environments for growing the new grasses
and different ways of using them. There
are several agencies more specifically
adapted than the Institution to take the
primary responsibility for such a program.
As the results of the proposed collaborative
tests become available, some agency other
than the Institution may desire to assume
complete responsibility for expediting their
practical evaluation and for introducing
the new grasses into commercial use.

Chromosome counts of numerous indi-
viduals of seven of the more interesting
double and quadruple Poa hybrids have
thrown further light on the mechanism of
inheritance in the apomictic grasses.

The genetics, cytology, and ecology of
the various races of single species which
are adapted to different climates have long
been of interest to the Department. New
hybrids are being made of various climatic
races of Mimulus, a group especially favor-
able for the analysis of the factors that
determine the fitness of different ecological
races to the climates for which they are
specifically adapted.

Studies by Drs. Ralph W. Chaney and
R. N. Lakhanpal of two fossil floras from
Oregon show that there were local differ-
ences in vegetation during the Oligocene
epoch similar to the differences existing
today on opposite sides of the Cascade
Range. It has become apparent that some
floral differences must be explained on the
basis of environmental dissimilarities rather
than of discrepancies in age, in order to
avoid errors in the dating of fossil-bearing
beds.

PERSONNEL

Biochemical Investigations

Staff: C. Stacy French, Director, Harold W.

Milner, James H. C. Smith, Herman A.

Spoehr, Chairman Emeritus, Violet K.

Young.
Visiting Investigators: Robert Hill, Jack

Myers, Hiroshi Tamiya, Hemming I.

Virgin.
Research Fellows: L. N. M. Duysens, Bessel

Kok, Robert W. Krauss.
Research Assistants: Allen Benitez, Edwin A.

Davis.
Technical Assistants: Richard M. Cook, Wal-
ter W. Holt, Elwin W. Seeley.

Experimental Taxonomy

Staff: Jens C. Clausen, Paul Grun, William

M. Hiesey.
Guest Investigator: Friedrich Ehrendorfer.
Stanford University graduate student associ-
ated with the Department: George H.

Ward.
Research Assistants: Charles T. Mason, Jr.,

Malcolm A. Nobs.
Technical Assistants: Robert W. Ayres,

Arthur L. Hawk, Edwin H. Ketchledge,

Edward L. Triplett.
Clerical Assistants: Marion S. Bracken, John

M. Bracken.
Gardener: Wesley B. Justice.

DEPARTMENT OF PLANT BIOLOGY

I 49

Research Associate

Ralph W. Chaney, Professor of Paleontology,
University of California, Berkeley.

Department Secretary
Wilbur A. Pestell.

Mechanical Engineer
Louis R. Kruger.

Custodians
Donald B. Muir, Richard P. Ludolph, Jr.

Dr. Paul Grun has been at the Karo
linska Institute, Stockholm, Sweden, since
September 1952, investigating the ultra-
violet absorption spectra of isolated root-
tip nuclei.

Professor Hiroshi Tamiya, Director of
Tokugawa Institute and Professor of Biol-
ogy of Tokyo University, was a guest of the
Department for several months while in
this country on a grant from the National

Research Council. During this time he
wrote two chapters for the Institution's
monograph Algal Culture: From Labo-
ratory to Pilot Plant and, with Mrs. Ta-
miya, prepared various foods containing
Chlorella which were evaluated for palata-
bility by the staff and by various visitors.

We were also fortunate this year to have
a visit of several weeks from Dr. Robert
Hill, of Cambridge University. This visit
provided a stimulating opportunity for
collaboration on several aspects of the
problem of the state of chloroplast pig-
ments in living cells, the results of which
are reported below under "The absorption
and fluorescence spectra of natural pro-
tochlorophyll."

Dr. Glenn W. Todd, of the California
Institute of Technology, visited the Depart-
ment for a week to measure the fluores-
cence spectra of the pigments of certain
light-sensitive seeds.

BIOCHEMICAL INVESTIGATIONS

The Protochlorophyll-Chlorophyll

Transformation : The Nature of

Protochlorophyll in Leaves

James H. C. Smith and Allen Benitez

The driving power of photosynthesis is
the light captured by chlorophyll. How
chlorophyll passes on this captured energy
to other parts of the photosynthetic system
has been the subject of a great deal of re-
search and speculation. Most of the earlier
conceptions of this transfer represented it
as a reversible chemical change involving
the pigment itself, but the more recent in-
terpretations have tended more and more
to ascribe it to physical phenomena. These
physical interpretations assume that the
pigment molecule undergoes no chemical
change, but because of its excited state in-
duces chemical change in other parts of
the photosynthetic system.

A factor that has contributed to lack of
clarity in regard to this energy transfer is
ignorance concerning the condition of
chlorophyll in the plant. Spectroscopic
examination has shown that chlorophyll
when in the plant exists in a different state
from that found when it is dissolved in
organic solvents. Until the reason for this
difference is clarified, it may be difficult to
ascertain whether the energy transfer in
photosynthesis is accomplished by chemical
or by physical means.

If chlorophyll in its natural state is
bound to a carrier, the resulting complex
can be considered to be a compound — pre-
viously named chlorophyll holochrome —
in which reversible chemical transforma-
tions could take place in the carrier portion
without causing a detectable change in the
pigment portion.

Various attempts have been made to iso-

150

CARNEGIE INSTITUTION OF WASHINGTON

late and determine the chemical nature of
the chlorophyll holochrome, but the re-
sults have been ambiguous. Much of the
ambiguity has come from inability to test
the isolated holochrome for unimpaired
function.

It seems likely that one approach to an
understanding of the nature of the chloro-
phyll holochrome is through an under-
standing of the protochlorophyll holo-
chrome. In the transformation of proto-
chlorophyll to chlorophyll, the chemical
change is small. Since the transformation
is so readily effected, it is assumed that the
change in the carrier portion of the holo-
chrome is also small. If, therefore, the
constitution of the protochlorophyll holo-
chrome could be determined, our knowl-
edge of the chlorophyll holochrome should
be greatly increased.

With this in mind, the objective of this
year's work has been to determine the na-
ture of the protochlorophyll holochrome by
examining the effects of various external
factors on the protochlorophyll-chlorophyll
transformation within the leaf, and by iso-
lating this holochrome in a state in which
it can be transformed. If the isolated holo-
chrome is active in respect to this trans-
formation, it may also show activity in
other respects related to photosynthesis.

Effect of temperature. The effect of tem-
perature on the protochlorophyll-chloro-
phyll transformation in the range from
+ 55.3 to -195° C was studied in detail.

When dark-grown barley leaves were
maintained at 39° C and above, they lost
their ability to carry out the transforma-
tion. The higher the temperature, the
more quickly they lost this ability. At
39° C the loss was barely perceptible after
an hour, whereas at 55.3° C it was 96 per
cent complete in 5 minutes. The capacity
for transformation was not lost on account
of destruction of the pigment portion of
the holochrome, since the unaltered pig-

ment could be extracted after the tempera-
ture treatment, but it was lost in all like-
lihood because of the effect of temperature
on the carrier. The pattern of behavior at
temperatures between 39° and 55° C
strongly suggests that the protochlorophyll
holochrome is a pigment-protein complex.

The extent of transformation of proto-
chlorophyll into chlorophyll for a definite
amount of radiation was measured at
several temperatures from room tempera-
ture to liquid-nitrogen temperature. As the
temperature was decreased, the percentage
transformation was decreased. The trans-
formation was 90 per cent at 23 ° C, 72 per
cent at — 20 °C, 36 per cent at —77° C, and
zero at —195° C. It is remarkable that
this biochemical transformation proceeds
to so great an extent at such a low tempera-
ture as —77° C. This provides evidence
against enzymatic control of the trans-
formation, because those enzyme systems
which have been studied at low tempera-
tures have shown no activity at —77° C,
and only slight activity at —20° C.

Freezing of the leaves at — io° and
— 20° C did greater damage to the trans-
formation system than was consistent with
the over-all trend of transformation with
temperature. Damage to biological sys-
tems by slow freezing is common, and
undoubtedly the damage to the protochlo-
rophyll holochrome can be ascribed to the
same cause. This interpretation is sub-
stantiated by the observation that leaves
frozen at much lower temperatures and
then brought to — 10 ° or —20° C some
time before being illuminated transformed
a greater percentage of their protochloro-
phyll than did leaves which had been ex-
posed at —io° or —20° C only.

Even when the illumination periods
were long, the transformation never
reached completion, but approached a lim-
iting value. This value was less, the lower
the temperature. At 20° C the limit of the

DEPARTMENT OF PLANT BIOLOGY

151

conversion of the protochlorophyll con- The fact that the reaction velocity agrees
tained in the leaves was 95 per cent; at with the second-order rate law and is pro-
— 40° C, 71 per cent; and at —70° C, 58 portional to the intensity of the radiation
per cent. This effect of temperature was suggests that the transformation is pro-
reversible; that is, leaves which had been duced either by the interaction of two
treated at a very low temperature in the protochlorophyll molecules, one of which
dark, when brought to a higher tempera- has been photoactivated, or by interaction
ture and illuminated, transformed the of a protochlorophyll molecule with a sub-
same percentage of their protochlorophyll stance formed by light absorbed in another
as if they had been treated at the higher protochlorophyll molecule,
temperature only. This was true so long The rate constant of this second-order
as the higher temperature was below the reaction was directly proportional to the
thawing point. If, however, the leaves light intensity and varied with the temper-
were first illuminated at the lower temper- ature. The values of the rate constant
ature and then at the higher temperature, ( X io 5 ) found at the various temperatures
the reversibility was only partial. In this were: 13.2 at 20° C, 7.4 at —40° C, and
case, the percentage transformation was 1.4 at —70° C. (The wave length of the
slightly below that obtained at the higher incident light used for these measurements
temperature, but it was considerably above was 579/577 mu.) The temperature coefn-
that obtained at the lower temperature, cient of the rate constant was what would
Thus it appears that illumination of the be expected for a photochemical reaction
holochrome, besides bringing about trans- that derived its activation energy almost
formation of protochlorophyll to chloro- exclusively from the light absorbed,
phyll, destroys to some extent the capacity These facts imply that native protochlo-
of the holochrome for transformation. rophyll is a pigment-protein complex whose

The progress of the transformation was phototransformation requires in some way

measured as a function of the amount of interaction between two protochlorophyll

radiation incident on the leaves. At molecules.

6° ±2° C, these measurements were made Attempted isolation of the holochrome.

at wave lengths 589, 579/577, 546, and 436 The second means used to determine the

mu. At the three longest wave lengths, the nature of native protochlorophyll was the

rate of the transformation was proportional isolation of the holochrome. In Year Book

to the square of the protochlorophyll con- No. 51 (1951-1952) it was reported that

centration, this second-order rate law be- etiolated leaves which had been disinte-

ing followed very closely. At 436 mu, grated by grinding in glycerine trans-

however, this law was not strictly obeyed, formed a large percentage of their proto-

The cause of the difference between the chlorophyll to chlorophyll when the brei

reaction at 436 mu and that at the other was illuminated. During the past year,

wave lengths is unknown. glycerine suspensions of ground dark-

If this were a strictly photochemical re- grown barley leaves have been obtained

action, in which one protochlorophyll mol- which had interesting properties. To pre-

ecule acted as the absorber of the activating pare the suspensions, the leaves were

light and also as the reactant, the conver- ground in glycerine and the liquid

sion would have progressed in proportion squeezed through a finely woven cloth,

to the amount of protochlorophyll present, The suspensions were freed of the largest

that is, according to the first-order rate law. particles by high-speed centrifugation.

152

CARNEGIE INSTITUTION OF WASHINGTON

Transparent, opalescent suspensions were mu. The dissimilarity in absorption by

obtained by this treatment. the chlorophyll holochromes — one newly

Such a suspension was found to contain formed from protochlorophyll holochrome,

an absorption band at 650 mu. When this the other from green leaves — suggests that

suspension was allowed to stand in the the state of native chlorophyll changes

dark at room temperature, the position of during its development and accumulation,

the absorption band shifted to 635 mu. A comparable observation has been made

The shift in position of the band was much by Krasnovskii.

slower at 6° C than at room temperature. Whole leaves, when frozen and thawed,

These facts have an important bearing on lose their capacity for photochemically

previous observations: The action spec- transforming protochlorophyll to chloro-

trum for the conversion of protochloro- phyll. The protochlorophyll holochrome in

phyll to chlorophyll in corn leaves had its glycerine suspension, however, after being

long-wave-length maximum at 650 mu, but kept for 3 hours at the temperature of solid

the fluorescence spectrum of dark-grown carbon dioxide and then brought to room

leaves had a maximum at 638 mu. These temperature, still retained this ability. This

two values were discrepant, since the fluo- is an interesting illustration of glycerine's

rescence band should have been at longer efficacy in preserving physiological activity

wave lengths than the corresponding ab- in biological material at low temperature,

sorption band. This discrepancy raised This effect may be useful in future work

troublesome questions concerning the in- on the isolation of the holochrome.
terpretation of the action-spectrum data. Etiolated leaves when ground with 62.5

But discovery that a freshly prepared glyc- per cent sucrose solutions also form sus-

erine extract of dark-grown leaves had a pensions which maintain the capacity of

maximum at 650 mu, identical in position protochlorophyll holochrome for photo-

with the action-spectrum maximum, and transformation to chlorophyll holochrome.

that this band shifted spontaneously to By expelling the strained and centrifuged

635 mu, a position consistent with the suspensions through a small orifice under

fluorescence band at 638 mu, gave a plausi- high pressure (see Milner et ah, Year Book

ble explanation for the discrepancy pre- No. 48, 1948-1949, p. 88) the particles

viously encountered. were reduced to very small size. Some

When the glycerine suspension was ir- particles, however, were still microscopi-

radiated, the protochlorophyll holochrome cally detectable with dark-field illumina-

absorption band at 650 mu decreased and tion. Whether these small particles were

a new absorption band with wave-length the seat of the transformation is yet to be

maximum at about 670 to 675 mu ap- determined.

peared. This new band was ascribable to The destruction of the protochlorophyll
the chlorophyll a holochrome. These facts holochrome, during the isolation process,
demonstrated that the protochlorophyll occurs more readily in some leaves than in
holochrome could be separated from the others: in etiolated barley leaves the de-
leaf in active form by the procedure de- struction is very rapid, whereas in etiolated
scribed. bean leaves it is relatively slow. Dark-
It should be noted that the position of grown barley leaves, ground with dilute
the absorption band of the chlorophyll salt solutions, lose nearly all their proto-
holochrome obtained from green leaves chlorophyll and fail completely to trans-
by grinding with glycerine was at 678 form what remains; etiolated bean leaves,

DEPARTMENT OF PLANT BIOLOGY

153

treated similarly, retain protochlorophyll
and the power to transform it.

Although a suspension of the proto-
chlorophyll holochrome can be obtained
from bean leaves by grinding with dilute
salt solutions alone, suspensions more satis-
factory in respect to stability, retention of
pigment, and power of transformation can
be obtained by grinding with glycerine or
concentrated sucrose solution. The position
of the bean protochlorophyll-holochrome
absorption maximum lies at 635 to 640 mu.
This position differs from that found for
the freshly prepared glycerine suspensions
of barley-leaf protochlorophyll holochrome,
650 mu, but agrees with that of prepara-
tions which have been permitted to stand
for some time at room temperature in the
dark. The significance of these observa-
tions is not yet clear, but it may be that
different protochlorophyll holochromes
exist.

Many experiments have been performed
to isolate the active protochlorophyll holo-
chrome from suspensions of it in salt solu-
tions, glycerine, and concentrated sucrose
solutions. As yet no real success has been
attained. The results obtained so far point
to the holochrome's being particulate, and
experiments are being planned to test this
assumption.

The Absorption and Fluorescence
Spectra of Natural Proto-
chlorophyll

Robert Hill, James H. C. Smith, and
C. S. French

The action spectrum for the transforma-
tion of protochlorophyll to chlorophyll was
determined several years ago (see Smith,
Koski, and French, Year Book No. 48,
1948-1949, p. 91). When this action spec-
trum was compared with the absorption
spectrum of protochlorophyll dissolved in
organic solvents, there was little room for

doubt that in the transformation proto-
chlorophyll was the light-absorbing agent
as well as the reactant. But discrepancies
existed: the positions of the action-spec-
trum bands were displaced from those of
the absorption bands of protochlorophyll
in organic solvents; and the shape of the
action-spectrum curve differed from the
shape of the absorption-spectrum curve of
protochlorophyll. These differences would
exist if the absorption spectrum of native
protochlorophyll were displaced from that
of the extracted pigment as is the case with
other plant pigments. More convincing
evidence than this assumption, however,
would be a demonstration that the absorp-
tion bands of natural protochlorophyll co-
incide with the bands of the action spec-
trum for its own transformation.

Experiments were performed to obtain
the absorption spectrum of dark-grown
barley leaves by use of a Zeiss microspec-
troscope. Difficulty was encountered, how-
ever, in seeing the absorption bands ini-
tially present. As soon as the light for
observing the bands was turned on, trans-
formation of the protochlorophyll began.
Rapid changes in absorption followed
which were exciting to watch but impossi-
ble to define. The impression was always
gained that originally a band existed near
650 mu which changed rapidly to a band
near 635 mu and that sometime during the
few seconds in which these fleeting changes
were occurring, the chlorophyll band near
670 mu appeared. This band continued to
increase in intensity until a sharp, stable
absorption band existed between 670 and
690 mu. The weak absorption band near
635 mu persisted.

In an attempt to stabilize the original ab-
sorption bands, etiolated barley leaves were
killed by immersion in hot water. After
this treatment there was no transformation
and a stable band existed in the leaves at
635 mu. This band was undoubtedly due

i54

CARNEGIE INSTITUTION OF WASHINGTON

to protochlorophyll, but was displaced
from the action-spectrum maximum at
650 mp. Because of this and other unac-
counted-for spectral changes, doubt was
cast on the heating procedure for maintain-
ing the spectrum of the living leaf.

Another tissue than etiolated leaves that
can be used for determining the absorption
spectrum of native protochlorophyll is the
inner coats of squash seeds. In many ways,
this material is ideally suited to this pur-
pose: it is translucent; it contains a high
concentration of protochlorophyll; it does
not transform protochlorophyll during rel-
atively long illumination periods; and it
is readily extracted with organic solvents
to yield protochlorophyll solutions.

Examination of inner coats from Hub-
bard squash seeds by means of a Zeiss
microspectroscope showed this tissue to
have absorption bands that agree closely
with the action spectrum for the transfor-
mation of protochlorophyll to chlorophyll
in etiolated corn leaves. The acetone ex-
tract of this tissue had absorption bands
which identified the extracted pigment
with protochlorophyll. The data substan-
tiating these statements are given in table 1 .
The similarity between the action spectrum
for the transformation of protochlorophyll
to chlorophyll and the absorption spectrum
of natural protochlorophyll is convincing
evidence for natural protochlorophyll's be-
ing the pigment active in the absorption of
light for the transformation reaction.

The fluorescence spectra of Hubbard
squash inner seed coats showed the ex-
pected protochlorophyll band at 655-660
mp, but there was from 2 to 10 times as
intense fluorescence in a band whose posi-
tion varied from 700 to 710 m|j. Further-
more, a band at about 638 m|j was also
present in the fluorescence spectra of ba-
nana squash inner seed coats and of
glycerine extracts of the inner seed coats of

Hubbard squash. Acetone extracts of all
seed coats, however, showed absorption
and fluorescence spectra identical with
those of protochlorophyll.

This state of affairs, though apparently
confusing, is probably merely the com-
bined result of two processes. One of these
is the change in situ of a part of the normal

TABLE 1

Summary of absorption maxima for the

pigment of the inner seed coats

of Hubbard squash

Wave lengths (m/i)

Action spectrum for the
transformation of
protochlorophyll in
corn leaves 650 593 548 445

Absorption bands of inner

seed coats 650 595 550 ...

Absorption bands of glyc-
erine suspension of
ground-up inner seed
coats (photoelectric) 650 590

Absorption bands of ace-
tone extract of inner
seed coats:

Visual 625 570 530 ...

Photoelectric 625 570 530 ...

Absorption spectrum of
acetone solution of
pure protochlorophyll
(photoelectric) 623 571 535 432

form of protochlorophyll, which absorbs
at 650 m\\ and fluoresces at slightly longer
wave lengths, to the altered form absorbing
at 635 mp. This efiect is presumed to be
responsible for the variable amount of
fluorescence at about 638 m^i. The very
strong fluorescence found in these deeply
colored seed coats at 700 to 710 mjj is
presumably due to a secondary band of
protochlorophyll. Its great intensity in
comparison with the "main" peak near 653
mp is believed to be due to the fact that
light of these wave lengths is not reab-

DEPARTMENT OF PLANT BIOLOGY

155

sorbed, whereas the "main" peak has been
nearly obliterated by internal reabsorption
because it is so close in position to the ab-
sorption peak of protochlorophyll. This is
indeed an extreme case of distortion due to
reabsorption. Perhaps an alternative ex-
planation based on the existence of an
unrecognized pigment fluorescing at 700
to 710 mu may be possible. No such pig-
ment is, however, extractable by acetone,
nor does the fluorescence of the acetone-
extracted residue show any sign of it.

The Fluorescence Spectra of
Chloroplast Pigments

C. S. French

Measurements of the fluorescence spec-
tra of red algae were made several years
ago specifically to study the transfer of
energy between their various chloroplast
pigments, as reported in Year Book No. 49
(1949-1950), page 86, and No. 50 (1950-
1951), page 122. A new fluorescent pig-
ment of leaves (Year Book No. 49, p. 99)
was found to be a porphyrin type of com-
pound by the characterization of its fluores-
cence spectrum. These and numerous un-
published determinations of the curve of
fluorescence intensity against wave length
as recorded photoelectrically have been
made for the purpose of settling specific
questions in regard to the constitution or
function of previously recognized pig-
ments.

During the course of these investigations
it has gradually become apparent how re-
markably little is known about the fluores-
cence spectroscopy of chloroplast pigments
in living material. Many of the curves that
have been measured are not clearly inter-
pretable without recourse to the entirely
unproved assumption of the existence of
previously unsuspected pigments or of un-
believably great distortion by internal ab-
sorption of the fluorescence light. Further-

more, the frequent absence of anticipated
fluorescence peaks in certain plants, for
example those of chlorophyll b or chloro-
phyll c, leads to the question whether this
absence is due to very efficient transfer of
energy from these pigments to others, or
to internal reabsorption of the fluorescence,
or merely to their low efficiency as fluores-
cence emitters.

In brief, it has been our experience that
the use of fluorescence spectroscopy to
answer a few questions has raised many
general problems of basic importance re-
lating to the interpretation of fluorescence
spectra. The use of fluorescence emission
curves in characterizing not only the na-
ture of the pigment systems in plants, but
more particularly the functional relations
between them, is a field that has hardly
begun to be developed to its potential value
for investigating photosynthetic pigments
in their natural state.

In order to develop the field of fluores-
cence spectroscopy of photosynthetic pig-
ments enough to make use of its inherent
possibilities, several different lines of in-
vestigation will have to be carried farther
and correlated with one another. In the
first place, fluorescence spectra of the com-
mon chloroplast pigments and of their
derivatives should be precisely determined
in the commonly used organic solvents.
Some of these spectra are available in the
literature and some have been measured in
this laboratory, but much routine work re-
mains to be done in compiling and check-
ing these reference data for pigment identi-
fication. In the second place, the fluores-
cence spectra of the individual pigments in
living organisms should be obtained. This
is the more difficult problem. The diffi-
culties here lie in the fact that except for
chlorophyll a and protochlorophyll, it is
hard to find organisms showing only the
spectrum of a single pigment. Further-

156

CARNEGIE INSTITUTION OF WASHINGTON

more, in order to eliminate errors due to ous substances in plants or in their extracts,

internal reabsorption, only plant materials does not as far as we know bear directly

containing low concentrations of the pig- on the nature and function of the pigments

ments are suitable. Many of these in vivo that are involved in photosynthesis,
curves for the individual pigments must The apparatus has been improved by the

be obtained by subtracting curves of installation of the recently available 6217

known pigments from curves obtained photomultiplier tube and the correspond-

from organisms containing two or more ing modifications of the correcting mecha-

fluorescent substances. A third phase of nism, so that the useful range now extends

the program is the exploration of a wider to 770 m|j rather than to the previous red

variety of plants that contain both simple limit of 750 mp. The sensitivity is also

and complex pigment systems. A fourth several times greater than before this

phase is concerned with establishment of change. The frequency of the tuned ampli-

suitable conditions and selection of appro- fier has been changed from 150 to 120

priate plant material for making the meas- cycles per second, and a rotating sector is

urements, and of suitable means for apply- no longer used for the fluorescence meas-

ing the necessary corrections. Such factors urements, since the light source itself is

as the well known change of fluorescence modulated at this frequency. A 120-cycle

intensity of chlorophyll with time in pho- chopper is, however, used with the stand-

tosynthesizing cells, and the limiting con- ard lamp for calibration,
centration of pigments to give curves During the past two years the develop-

undistorted by absorption, must be con- ments in this program have been of very

sidered. Furthermore, a single pigment limited scope in comparison with the po-

may exist in two different forms within tentialities of the subject, although a small

the cells, as was found to be true of pro- number of new phenomena have been en-

tochlorophyll. countered. One of the main questions in

In addition to the biological problems, mind has been to see why the expected

continual attention must be paid to the fluorescence bands of chlorophylls b and c

apparatus, to take advantage of recent com- do not appear in the spectra. The fluo-

mercial developments in optical, electronic, rescence of chlorophyll b has been reported

and intensity-measuring devices as well as in the older literature in photographic fluo-

high-intensity light sources. Since the fluo- rescence spectra of several plants. The

rescence light emitted by living plants is absence of fluorescence of chlorophylls b

very weak, even small improvements in the and c in our experiments may well be due

equipment often have a significant effect to the nearly complete transfer of energy

on the results. to chlorophyll a, which process may per-

The program is not being followed in haps proceed with an efficiency dependent

the order here given, but this outline is on the good health of the plant material

used as a logical framework into which used. We have looked for chlorophyll b

may be fitted the various pieces of infor- fluorescence in very pale leaves of various

mation acquired in the course of experi- species, in the hope of avoiding possible

mentation carried out for many different difficulties due to reabsorption of the fluo-

shorter-term objectives. We have avoided rescence, which should, owing to the posi-

the entire field of nonchloroplast pigment tions of the absorption bands, be greater

fluorescence, which, though it provides re- for chlorophyll b than for chlorophyll a.

markably sensitive means of detecting vari- These measurements have not led to the

DEPARTMENT OF PLANT BIOLOGY

157

detection of chlorophyll b fluorescence, but
they have given with greater precision the
shape of the fluorescence spectrum of chlo-
rophyll a in its native state. The compila-
tion of such basic data is essential for
further progress and has in itself been one
of our immediate aims.

Chlorophyll c is found in diatoms and
certain algae along with chlorophyll a.
A culture of the diatom Nitzschia Clostrid-
ium minutissima was kindly given to us
by Professor C. B. van Niel. Measurement
of its fluorescence spectrum confirmed the
earlier finding of other investigators that
this organism does not show the band of
chlorophyll c at about 640 mp, where it
would be expected on the basis of its ab-
sorption spectrum in organic solvents. In
addition to the chlorophyll a fluorescence
spectrum, these diatoms did, however, have
a peak of 705 mu. At present this is pre-
sumed to be the long-wave-length second-
ary band of chlorophyll c fluorescence,
and the absence of the main band is
attributed to its reabsorption by chloro-
phyll a within the cells. Even extremely
dilute suspensions of the diatoms show
this behavior, so the reabsorption, if it
actually is the cause of the band disappear-
ance, must take place within the individual
cells. The effectiveness of different wave
lengths in producing chlorophyll a fluo-
rescence in Nitzschia showed that fuco-
xanthin, which has its maximum in vivo
absorption at about 510 mu, is more effec-
tive in transferring energy to chlorophyll
than are the carotenoids, which absorb
from 400 to 450 mu.

It should be noted that in the region
700-710 mu the positions of the fluores-
cence bands from the diatom N. Clostrid-
ium minutissima and from inner seed
coats of squash are very nearly the same.
These biological materials contain chloro-
phyll c and protochlorophyll respectively.
Inasmuch as the absorption spectra of these

pigments when extracted are similar, as
well as their in vivo fluorescence spectra,
an interesting question is raised concern-
ing the relationship between chlorophyll c
and protochlorophyll.

A nearly complete obscuring of the main
chlorophyll fluorescence peak was found in
a deep green leaf, Photinia arbutijolia, il-
luminated with green light (546 mu)
which penetrates the leaf deeply. With
blue light, however, 436 mu, absorbed in
the outer layers of the leaf, the main peak
and the secondary peak were of equal
height. These measurements, made in col-
laboration with Dr. Robert Hill, were later
confirmed by Dr. Young. This effect is
an excellent example of the difficulty of
clearly recognizing the effects of reabsorp-
tion when using data which taken at their
face value would appear to indicate that
all the fluorescence present originated from
an unknown pigment.

The fluorescence spectra of a normal red
alga were compared with those from the
same alga adapted to red light, in col-
laboration with Professor Lawrence Blinks
and Mr. Robert Airth. The red-adapted
alga gave less fluorescence than the normal
one in the region around 700 mu, for some
reason that is not yet clear.

Dr. Glenn W. Todd measured the fluo-
rescence spectra of a pigment isolated from
light-sensitive bluegrass seed, and found
this pigment to be identical with pheophy-
tin as to its absorption and fluorescence
spectra.

Reversible Changes in the Light Absorp-
tion of Purple Bacteria Caused
by Illumination

L. N. M. Duysens

Some intermediate products of photo-
synthesis have a natural lifetime shorter
than a few seconds. The elucidation of
the nature and function of these inter-

!^8 CARNEGIE INSTITUTION OF WASHINGTON

mediates would probably contribute greatly their growth medium — i per cent peptone,

to an understanding of the photosynthetic 0.5 per cent sodium chloride — the change

reactions. It is not feasible to study inter- was zero at low intensities from about

mediates with such a short life by chemical 10 4 ergs/cm 2 /sec, and increased slowly up

analysis after extraction. Rather, it is neces- to the highest intensity used, 9 X io 4 ergs/

sary to work with living cells, or parts of cm 2 /sec. In this medium, even at the

cells able to produce intermediates upon highest intensity used, the maximum

illumination. These intermediates may be change measured was smaller than the

detected by observing the effects produced greatest change obtained in distilled water,

by irradiation. Two such effects are fluo- The maximum change in absorption coeffi-

rescence and luminescence of photosyn- cient at 880 mjj was of the order of 1 per

thetic pigments. cent.

A third effect, a change in the absorp- If we assume that the rate of formation

tion spectrum of a photosynthetic pigment, of the changed state in the light is the

has been observed. This effect was found same in the presence and in the absence of

in the Biophysical Research Group at peptone, then the fact that the change ob-

Utrecht during investigations on the trans- served is smaller in peptone than in dis-

fer of excitation energy between pigments tilled water shows that the rate of disap-

in photosynthesizing cells. Such a change pearance of the changed state in dark

in absorption indicates a change in the pig- reactions must be higher in peptone than

ment or in adjacent cell constituents. A in water. The high rate of disappearance

study of this change may throw light on in the presence of peptone may be caused

the reactions following excitation of chlo- by the reaction of the changed pigment or

rophyll a in algae or bacteriochlorophyll of a substance causing the changed absorp-

in bacteria. tion with a product formed from peptone.

It was found that irradiation of purple This reaction is perhaps a photosynthetic

bacteria with blue light of high intensity one, since peptone is the substrate used by

caused a change in their absorption spec- the bacteria for growth in the light. The

trum. The change was completed within study of the formation and disappearance

one second. When the cells were darkened, of such an intermediate presumably would

they reacquired their original absorption provide data important for the elucidation

spectrum, also within one second. The of the mechanism of photosynthesis. A

change in absorption in the region 650 to first step in such a study is the measure-

920 mn indicated that the bacteriochloro- ment of the rate of disappearance of the

phyll absorption was altered by irradiation, changed state both in the presence and

The change in the absorption coefficient at in the absence of peptone. For this pur-

880 mu, the wave length at which the pose the following experiments have been

effect was maximum, was measured in the carried out at the Department of Plant

nonsulfur purple bacterium Rhodospiril- Biology.

lum rubrum strain 4 as a function of the The bacterial suspension was circulated

intensity of irradiation. If the bacteria in a closed system by means of a centrif-

were suspended in distilled water, the ugal pump. It was illuminated while be-

change increased roughly linearly with ing passed through a coil of transparent

incident intensity until saturation was Tygon tubing, 100 cm long. It then passed

reached at about 2X10 4 ergs/cm 2 /sec. If, in darkness to an absorption cell where

however, the bacteria were suspended in the changed absorption caused by illumi-

DEPARTMENT OF PLANT BIOLOGY

159

nation could be observed. The distance Table 2 shows the changes in absorption
from the end of the coil to the mid-point of for short dark path measured after bub-
the absorption cell was normally 11 cm. bling the designated gas mixture through
This distance in darkness could be length- the suspension for about half an hour. The
ened to 33 cm by covering part of the coil, changes were roughly reversible.
From comparison of the measurements ob- The table also shows half lives of the
tained after different durations of dark changed state, calculated on the arbitrary
period with the measurements obtained assumption that the restoring reaction is
without illumination of the coil, and from proportional to the number of changed
the rate of flow of the suspension, 45 cm/ molecules. The fact that the half lives are
sec, the rate of return of the changed ab- essentially the same suggests that the re-
sorption to "normal" could be computed, storing reactions occur with the same rates
This rate was taken as a measure of the for the two oxygen concentrations, al-
rate of the restoring reactions.

The light used for the absorption meas- TABLE 2

urements, wave length 880 mu, was ob- Influence of oxygen on change in absorption

tained by use of a tungsten lamp and a of purple bacteria when illuminated

monochromator, and was of such low in- =====^^====^=^==

tensity as to have no effect on the absorp- Per cent oxygen ^^ ch H a ^ fe st a f te

tion. The temperature of the reacting sys- in y rogen absorption (sec)

tern was 30 ° C. ~ 08 ~

If the intensity of the light incident on 4 0.32

the bacteria in the coil was increased, the 14 0.45 1.0

measured change increased to a limit. In 43 °-? 8 * - 2

all experiments the intensity of the inci-
dent light used was great enough to obtain though the changes in absorption are dif-
this limit. The change in absorption is ferent.

calculated as the percentage change in ab- To investigate the nature of the action

sorption coefficient at 880 mu. of oxygen, the change in absorption by a

The same strain of Rhodospirillum ru- colloidal aqueous suspension of the bac-
brum that was used in Utrecht, grown in terial material was measured. This ma-
1 per cent peptone and 0.5 per cent sodium terial was prepared by passing a concen-
chloride, was used throughout the experi- trated suspension of bacteria at 1400 at-
ments. It was observed that after centrifu- mospheres pressure through a slightly
gation and resuspension of the bacteria in opened needle valve (Year Book No. 48,
0.5 per cent sodium chloride and M/15 1948-1949, p. 88). The larger particles,
phosphate buffer of pH 6.5, the change in such as cell walls, were removed by cen-
absorption gradually declined. The original trifugation for 15 minutes at 10 4 times
change in absorption was restored if the gravity. The clear colloidal suspension
suspension was removed from the circuit showed an appreciable change in absorp-
and permitted to stand in contact with air tion, 1.78 per cent, which was reduced
for some time. This observation suggested only 30 per cent by sweeping with hydro-
that the effect was caused by oxygen. Vari- gen for 18 minutes. The original change
ous mixtures of hydrogen and oxygen were in absorption was restored either by vigor-
bubbled through part of the circuit before ously bubbling oxygen through the sus-
and during the absorption measurements, pension or by making it 7X10" 3 M with

i6o

CARNEGIE INSTITUTION OF WASHINGTON

potassium ferricyanide as oxidant. The
phenomena were not influenced by the
presence of 3X10" 4 M potassium cyanide
or 0.1 M urethane.

If one-day-old bacteria are centrifuged
and resuspended in fresh growth medium,
in which they may be expected to photo-
synthesize, their change in absorption is
negligibly small up to irradiation intensi-
ties of io 5 ergs/cm 2 /sec. This intensity re-
fers to the fraction of irradiation of wave
lengths shorter than 900 mu, where the
bacteria show strong absorption. On the
other hand, if bacteria or their extracts are
in such a condition that they show a meas-
urable absorption change, this change
reaches its maximum at intensities between
5X10 3 and 5X10 4 ergs/cm 2 /sec. Under
these conditions the half lives of the
changed states in distilled water with 0.5
per cent sodium chloride were found to be
substantially alike in the presence of hydro-
gen, nitrogen, 2 per cent carbon dioxide,
and helium. The change in absorption of
bacterial suspensions was greater at pH 8.0
than at pH 6.5, and greater without sodium
chloride than with 0.5 per cent sodium
chloride.

The ratio of actinic intensities of blue
and infrared radiation just sufficient to
saturate the change in absorption was
roughly equal to the ratio that was to be ex-
pected if radiation energy absorbed by or
transferred to bacteriochlorophyll was re-
sponsible for the change.

Our observations can be explained by
the assumption that a change in absorp-
tion upon irradiation is observable if a
certain substance in its oxidized state is
present in the bacterial system. If this
substance is reduced, the change in absorp-
tion becomes too small for measurement
with our apparatus. It is known that bac-
teria respire vigorously in a fresh suspen-
sion supplied with growth medium. They
may well be able to remove the oxygen and

bring about the reduction of the oxidized
substance. This explains why a change in
absorption is not found under these con-
ditions. If a change in absorption is not
observable with our apparatus, the changed
state may nevertheless be produced but
with such a short life that the original ab-
sorption is restored before the suspension
reaches the absorption cell.

The Effect of Light on the Consistency
of Cytoplasm

Hemming I. Virgin

Visible light has many different effects
on normal green plants. Several more or
less independent light-influenced processes
are simultaneously going on in the plant
cell, each having its own characteristic
action spectrum depending on the pig-
ment which is active in the process. Some
of these processes, for example photosyn-
thesis, require rather high light intensities;
others, such as phototropism and photo-
periodism, show great sensitivity to very
low intensities. The reactions induced by
low light intensities undoubtedly play a
role in maintaining the orientation and
shape of the living plant.

Some years ago it was found that the
cytoplasmic viscosity of cells of Elodea
densa and Spirogyra sp. as determined by
the centrifuge method fluctuates under
natural conditions. The fluctuations are
caused by light and disappear after the
cells stay 3 days in darkness. The cyto-
plasm now shows a very high response to
light, consisting in a rapid decrease of vis-
cosity when illuminated with strong inten-
sities, and a slow increase with weak in-
tensities which can be far below those
required for photosynthesis.

The work being carried on at the De-
partment of Plant Biology has two main
objectives: first, to determine whether this
phenomenon is restricted to a few special

DEPARTMENT OF PLANT BIOLOGY

161

plant species or whether it is widespread
among plants; second, to identify the pig-
ment responsible for this process by means
of action-spectrum measurements.

The cytoplasm of water plants has a
much lower viscosity than that of land
plants; consequently a much smaller cen-
trifugal force is required to displace its
contents. In order to work with terrestrial
plants, a centrifuge more powerful than
the one used previously was constructed.
With a force of 12,500 times gravity the
cell contents of most land plants are com-
pletely displaced within 1 to 4 minutes.
During the centrifugation, the plant tis-
sues are placed in small glass tubes meas-
uring 15X2 mm inside. The centrifuge
attains maximal speed in 4 seconds and
can be stopped in an equal time. A short
braking time is necessary to make observa-
tion possible before redistribution of the
cell contents takes place.

With this centrifuge it has been possible
to establish the fact that light has a definite
effect on the cytoplasm of terrestrial plants,
for example on the leaves of Eschscholtzia
and on the petioles of Taraxacum. In
order to have the cytoplasm respond maxi-
mally, the plants must be kept either in
very weak light or in darkness for a few
days. These findings make it likely that
the effect of light on protoplasmic viscosity
occurs in many or all plants. Its role, from
a physiological point of view, has not been
found.

The experiments concerning the action
spectrum for this phenomenon have been
performed with the Wood grating mono-
chromator, using a very high-pressure mer-
cury arc as light source. The results ob-
tained have confirmed the earlier prelim-
inary results found with interference filters,
namely, that the process is activated mainly
by blue light. The effect is restricted to
wave lengths from about 400 to 500 mu,
with peaks of activity at 450 and 480 mu.

Exploratory experiments have failed to
show any response to infrared radiation.
The spectral region of greatest activity
corresponds to that absorbed most strongly
by the carotenoids and by riboflavin-like
pigments. It is possible, therefore, that the
pigment active in changing the viscosity of
protoplasm belongs to one of these classes,
which have been postulated to play a role
in the better-known phototropic response,
to which this phenomenon may be related.

Up to the present time, the displacement
of the cell contents has been determined by
visual microscopic examination of the plant
tissue. In the hope that a more objective
method of measurement might be devised,
the light transmission of illuminated and
centrifuged Elodea leaves was measured
and compared with that of unilluminated
leaves. The transmission was changed
only a few per cent by the treatment of
the leaves. Consequently, this method of
measurement did not seem promising for
practical use.

The possibility exists that the light-in-
duced changes in the viscosity of the cyto-
plasm may affect the rigidity of the irra-
diated tissue. To test this possibility, a
method of measurement was suggested by
Dr. Bush which is based on the principle
that the oscillation frequency of a vibrat-
ing rod depends on its rigidity. In the ex-
perimental set-up a small piece of plant
tissue was fastened at one end between two
glass plates. To the other end was attached
a very small steel plate. The tissue was
placed in front of an electromagnet fed
with alternating current at various fre-
quencies from an oscillator. If the turgor
of the tissue was altered by dipping it in
sucrose or salt solutions of various concen-
trations, significant changes occurred in
the frequency which gave the maximum
vibration. But when the rigidities of il-
luminated and unilluminated tissues were
measured, no difference was found be-

1 62

CARNEGIE INSTITUTION OF WASHINGTON

tween the two. The light-induced altera-
tion in consistency of the cytoplasm is
therefore probably very small in compari-
son with the rigidity of the leaf, which is
determined mainly by the turgor pressure
of the cytoplasm on the cell walls. This
experimental procedure may perhaps be of
use in the investigation of osmotic effects
in living material.

The Curve Analyzer

C. S. French, W. R. Fair, R. M. Cook,
W. W. Holt, and Elwin W. Seeley

The machine for converting curves and
plotting equations (Year Book No. 49,
1949-1950, p. 100; No. 50, 1950-1951, p. 133;
No. 51, 1951-1952, p. 154) that has been
under construction since January 1950 has
been developed to the point of reasonably
reliable operation, and no further extensive
expansion or alterations are contemplated
at present. Its functional units now consist
of a voltage source increasing linearly with
time, which is used as the independent
variable, and of five curve followers, nine
adding amplifiers, three amplifiers used
for curve multiplication, two integrators,
and a recorder. There are also power sup-
plies and amplifiers necessary for the oper-
ation of the fourteen servo systems driving
the tables, followers, integrators, and re-
corder. Since one integrator shares the
servo amplifiers of one follower, only four
followers can be used if both integrators
are required. The operational units may be
connected in any desired way for the trans-
formation or analysis of curves or as an
analogue computer for plotting the solu-
tion of equations.

A general view is shown in plate 1. In
this picture the rack at the left contains the
power supplies, the next one is a curve-fol-
lower unit with its table at the bottom
position, and the second follower unit,
bearing the curve y — e~ x , has the table at

the top of its range. The next rack, with
the white base, is the recorder, which is
similar in construction to the curve fol-
lowers. On its right is the main control
rack, carrying the independent variable
voltage supply, the adding amplifiers, and
one integrator with its driving amplifiers.
It also has a small patching panel on the
side, so that certain analogue voltages may
be measured or their interconnections
changed from the operator's position, al-
though most of the connections are made
on the rear of the control rack. The three
curve followers to the right of the control
rack are shown with normal probability
curves on their tables in the way in which
they are used to synthesize a complex
curve of adjustable shape. Follower num-
ber three, to the right of the control rack,
also carries the second integrator just be-
low the bottom of the table travel. Both
integrator assemblies can be pulled out on
roller slides to make them accessible when

in use.

During the past year the last two fol-
lowers were completed, the second integra-
tor was installed, and most of the ampli-
fiers were modified. The major change,
which has greatly improved the perform-
ance with problems involving cascaded am-
plifiers, has been the incorporation of
trimming condensers to correct for slight
phase shift of the analogue voltage at each
amplifier input. The addition of means for
adjustable damping of the follower car-
riage servo systems and an improved set-
ting procedure has removed the need for
frequent readjustment. New paraphase
amplifiers were built for use in curve mul-
tiplication and are entirely satisfactory.

The device has been tested with a
variety of problems, mainly those that oc-
cur in connection with other work of the
laboratory, and has been used extensively
for the recalculation of recorded curves of

Department of Plant Biology

Plate t

H
Z

X
H

Q
W
H
W

►J

o
o

w
o
w
(A

w
H

o
u

w
P

o
►J

z
<

as
w

N
I*

P
u

w
H

DEPARTMENT OF PLANT BIOLOGY

l6 3

fluorescence spectra. A description of the
machine, including records of sample prob-
lems, is being prepared for publication.

The following persons rendered valuable
services in connection with the construc-
tion of the curve analyzer during the year:
Messrs. Arthur B. Dennis, Alfred M.
Faries, Euyen Gott, Louis H. Rorden, and
particularly Louis R. Kruger.

Growth of Chlorella in Intermittent
Light 1

J. Neal Phillips, Jr., and Jack Myers

Last year's report included several studies
concerned with turbulence and the inter-
mittent-light effect on growth of Chlorella
(Year Book No. 51, 1951-1952, pp. 137-
140). One of these studies has been ex-
tended and the practical and theoretical as-
pects of the data have been examined.

Attempts at mass culture of algae have
posed the problem of increasing the effi-
ciency of utilization of the high intensity
of sunlight for photosynthesis and growth.
Efficiency of utilization of high light in-
tensity by the photosynthetic process is
known to be increased when the light is
presented in very short flashes separated
by long dark periods. In a dense culture
under sunlight illumination, individual
cells are subjected to a regime of intermit-
tent illumination which depends on the
thickness and density of the suspension and
the degree of turbulence. The question
therefore arises as to the proper regime
of intermittence to be provided by turbu-
lence in order to take advantage of the
flashing-light effect.

A constant-density, constant-volume
method was devised for measurement of
the specific growth rate in a very dilute
algal suspension illuminated with light
from a tungsten lamp at a color tempera-

1 This work is being done at the University
of Texas.

ture of 2900 ° K and at a maximum inten-
sity of 25 X io 4 ergs/cm 2 /sec in a focused
optical system. The maximum intensity,
which was nearly twice that required for
saturation, could be decreased by use of
neutral niters to provide various intensities
of continuous illumination, or the light
beam could be chopped by a set of rotating
sectors to give light flashes of 1, 4, 16, or
67 milliseconds combined with various
dark periods.

For practical purposes the data may be
treated by comparing the growth rate pro-
duced by short flashes of high intensity
with the growth rate under an equivalent
amount of energy supplied continuously at
a lower intensity. The comparison pro-
vides a measure of the degree to which
Chlorella pyrenoidosa can integrate inten-
sity X time for any intermittent-light regi-
men. With 1 -millisecond light flashes the
integration is essentially complete, since for
any value of integrated intensity the
growth rates are the same in intermittent
as in continuous light. As the flash time is
increased from 1 to 4 to 16 to 67 milli-
seconds, the degree of integration becomes
progressively lower. It is significant, how-
ever, that even with a 67-millisecond flash
there is some integration and therefore
some gain in efficiency of utilization of
high light intensity. The data predict that
a dense algal culture growing under sun-
light illumination will experience a signifi-
cant increase in growth if the turbulence
is sufficient to move cells in and out of the
front surface at such a rate as to give flash
times between 0.001 and 0.1 sec.

For theoretical treatment, the specific
growth rate, \, may be converted to
^/flash, or to the equivalent unit of
mm 3 2 /mm 3 cells/flash used in manomet-
ric experiments of photosynthesis studies.
Yield per flash for any flash time may then
be plotted against length of the intervening

164

CARNEGIE INSTITUTION OF WASHINGTON

dark period. Classic flashing-light theory,
developed from the well known experi-
ments of Emerson and Arnold, predicts
that for flashes of saturating intensity the
yield per flash will increase with increasing
dark times and will approach a maximum
yield which is independent of any charac-
teristic of the flash itself. The present data
show that as the light flash is lengthened
from 1 to 67 milliseconds, the maximum
yield per flash and the dark period neces-
sary to attain maximum yield are continu-
ously increased. Furthermore, the maxi-
mum yields per flash and the necessary
dark periods are much greater than those
observed by Emerson and Arnold with
their 0.01 -millisecond flashes. The present
data are compatible with those of Tamiya,
obtained from studies on photosynthesis
with 1- to 8-millisecond flashes; they are
not compatible with classic flashing-light
theory even when corrections are made for
the amount of photosynthesis and growth
occurring during the appreciable length of
the light flashes. It is concluded that there
is some unrecognized limitation in the
classic flashing-light theory or that some
new phenomenon is introduced by the use
of longer light flashes.

Selection of Algae Adapted to Large-
Scale Algal Culture 2

Jack Myers

The problem of commercial large-scale
culture of algae has been approached as a
problem of applied photosynthesis, that is,
the utilization of sunlight in the produc-
tion of organic matter. Both laboratory
and pilot-plant experimentation have used
well known species of Chlorella or Scene-
desmus as the algal material. The problem
may be approached also as one of applied
microbiology. Attention then is focused on
the importance of obtaining different spe-

2 This work is being done at the University of
Texas.

cies or strains of algae that are advanta-
geous in their particular responses to en-
vironmental conditions or in their capac-
ities for production of special materials.
The possibilities, limitations, and accom-
plishments of the second line of approach
are the subject of this report.

Selection of algal strains for special re-
sponse to environmental conditions. For
mass-culture applications, the most impor-
tant environmental conditions are tem-
perature, light intensity, and composition
of culture medium.

An encouraging beginning in the selec-
tion of special algal strains was made by
Sorokin's isolation of a number of strains
of Chlorella with higher temperature op-
tima than those of the strains generally
used (Year Book No. 51, 1951-1952, p.
141). The question of the basic reasons for
the marked difference in temperature op-
tima, 25-26° C for the standard Chlorella
pyrenoidosa and 38-39° C for one of the
new strains labeled Tx 71 105, is now under
study. At 30° C the standard Chlorella
pyrenoidosa will not grow indefinitely, but
will show high and only slowly decreasing
rates of respiration and photosynthesis for
many hours. It is inferred that temperature
optima for growth are governed, not by
photosynthetic or respiratory processes per
se, but by some more sluggish mechanism
such as the balance between rate of forma-
tion and denaturation of enzyme systems.
The time characteristics of response of
metabolic processes to different tempera-
tures are being studied in comparing the
two strains.

The ease and success of Sorokin's iso-
lation of strains of Chlorella with high
temperature tolerance has led to similar at-
tempts with the blue-green algae. The
combined efforts of Mr. William A. Kratz,
of our laboratory in Texas, and Dr. Mary
Belle Allen, of the Hopkins Marine Sta-
tion, have led to isolation of a blue-green

DEPARTMENT OF PLANT BIOLOGY

I6 5

alga tentatively identified as Anacystis rates or physical characteristics which make
nidulans. At the optimum temperature, them useful for mass-culture applications,
tentatively established as somewhere be- A search for desirable nitrogen fixers is
tween 40° and 44° C, it has a growth rate now being made by setting up accumula-
comparable with that of Sorokin's high- tion cultures in nitrogen-free media from
temperature strain of Chlorella. Kratz is rice-field inoculations. Chlorella is unusu-
now engaged in a study of Anabaena, Ana- ally versatile in its tolerance of variation of
cystis, and other blue-green algae under a pH and salt concentration; other algae are
grant from the National Science Founda- known to have restricted tolerance limits,
tion. The information should prove useful It may be expected that many of the fresh-
in affording a background for the possible water planktonic forms will be restricted
mass culture of blue-green algae. to rather low salt concentrations. Little

The need for selection of algae particu- attention has been paid to the marine algae,

larly adapted to high light intensity is illus- although these would have obvious ad-

trated by a limitation experienced with the vantages for mass-culture applications,

standard Chlorella. In a dense culture the Selection of algal strains for the produc-

average light intensity per cell is low, even tion of special materials. The search for

under sunlight illumination. At this low algae having a particular response to an

average light intensity each cell produces environmental condition may be carried

so much chlorophyll that it absorbs some- out generally by conventional accumula-

where between 10 and 20 times more light tion-culture methods so designed as to give

than it can use efficiently when it is ex- a selective advantage to those algae having

posed momentarily to full sunlight. The the desired characteristic. In the search for

property of producing an abundance of algae producing special materials, however,

chlorophyll is a distinct disadvantage for there is seldom any selecting principle

purposes of mass culture. The most desir- which may be imposed. The procedure

able organism would be one which has the must be the laborious one of screening

same quantum efficiency as Chlorella at many algae, already isolated, for some

low light intensities, but so little pigment desirable feature of their composition,

that it would require full sunlight intensity Models for necessary procedures are pro-

to attain light saturation. There is, of vided on the one hand by the study of

course, no certainty that such an organism sterol content of various algae by Krauss

exists, but it might be sought among plank- (Year Book No. 51, 1951-1952, pp. 142-

tonic algae which in nature grow under *45)> and on tn e other hand by the inten-

full sunlight intensities. sive search for antibiotic-producing micro-

In selection of algae adapted to particu- or g ani sms now being carried on by the

lar nutritional requirements, the important fermentation industries,
considerations are the nitrogen source, pH,

and total salt concentration. Chlorella will SpECIES OF Algae Adaptable to Mass

I TTT TTTTJTT

utilize ammonia, nitrate, and urea as nitro-
gen sources, but not all algae are equally Robert W. Krauss and Donald H. Boughton
versatile. The most desirable organism Studies in various laboratories concern-
would be one capable of nitrogen fixation ing the growth of green algae in large-
as observed in certain blue-green algae; scale cultures have involved species of

unfortunately, the nitrogen fixers now * TMs work is being done at the University of

available in culture do not have growth Maryland.

i66

CARNEGIE INSTITUTION OF WASHINGTON

Chlorella as the principal test organisms.
The choice of Chlorella has been logical in
view of its well known high growth rate
and its capacity for altering its chemical
composition under different environmental
conditions. Other unicellular or colonial
species of algae may, however, prove equiv-
alent or superior to Chlorella for mass cul-
ture. In order to determine how widely
applicable mass-culture techniques may be,

solution plus micronutrients. A mixture
of 5 per cent C0 2 in air was supplied, with
tap water flowing through stainless steel
tubing to hold the temperature at satisfac-
tory levels. The algae were harvested by
centrifugation and stored in a deep freeze.
Table 3 gives the average daily fresh-
weight yield per liter of ten species of algae
cultured during the fall and early winter
of 1952. With the exception of Scenedes-

TABLE 3
Total harvest weight and yield of algae grown in mass culture

Total fresh wt.
(g)

Species and source

Yield, fresh wt.
(g/l/day)

Carboys

Vats

Carboys Vats

1,100

0.076

0.036

1,232

0.071

0.038

1,075

0.157

0.087

739

0.200

0.115

872

0.093

0.046

1,761

0.117

0.092

1,309

0.084

0.044

1,625

0.233

0.123

360

0.227
0.076

0.064

10,073

Ankistrodesmus Braunii (Pringsheim 202/7b) 715

Ankistrodesmus falcatus (Pringsheim 202/3) 720

Asterococcus superbus (Pringsheim 3 /3a) 906

Chlamydomonas applanata (Pringsheim 11/2) 1,010

Coccomyxa simplex (Pringsheim 216/9a) 637

Coelastrum proboscidium (Van Niel Z.4.1.1) 800

Palmellococcus miniatus (Van Niel Z.l 1.1.1) 631

Scenedesmus sp.* (Van Niel Z. 14.4.1) 1,285

Selenastrum minutum (Pringsheim 278/3) 1,719

Stichococcus bacillaris (Van Niel Z. 15. 1.1) 509

Totals 8,932

* Not identified.

and to produce significant quantities of mus sp. (Van Niel Z.14.4.1), cultured dur-

other species of algae for analytical pur- ing both periods, these data do not include

poses, a screening program for species yields for the species of algae reported

other than Chlorella has been conducted, earlier (Year Book No. 51).

The culture method and apparatus have Rigorous comparisons of growth in cul-

been described previously (Year Book No. tures subjected to daily and seasonal vari-

51, 1951-1952, pp. 142-145). The culture ations in light are not justified. In general,

vessels, consisting of three 400-liter vats however, the culture conditions were simi-

and thirty 18-liter carboys, were installed in lar and the yields are a valid comparative

a greenhouse and received only natural measure of performance. Ten dry-weight

illumination. Inocula were obtained from determinations indicated an average water

pure cultures maintained on agar slants.
Concentrated cultures were grown in Er-
lenmeyer flasks in the laboratory prior to
inoculation of the vats and carboys. All

content of 70 ±7 per cent for the frozen
samples of algae. This did not apply to
Selenastrum minutum, which came from
the centrifuge as a viscous liquid with a

cultures were grown in a modified Knop's high water content. The mucilaginous cell

DEPARTMENT OF PLANT BIOLOGY

167

wall appears responsible for the poor sep- we have clear-cut evidence of essentiality
aration from the medium. Attempts to for algal growth are iron, manganese, zinc,
obtain heavy inocula in flask culture copper, and calcium. Cobalt may be in-
proved unsuccessful for the following spe- eluded in the list by virtue of its incorpora-
cies: Dunaliella salina, Euglena gracilis tion in vitamin B12. Molybdenum also has
var. bacillaris, Haematococcus pluvialis, been proposed for certain algae, but
Navicula sp., Nitzschia jrustulum, Oocystis whether this requirement holds for all
Naeglii, Selenastrum bibraianum, and Tri- classes remains uncertain. Failure to sup-
bonema aequale. It is possible that media ply any one of the essential elements in
containing other than the strictly inorganic sufficient quantity either reduces growth
nutrients employed in these experiments or halts it. The problem of supplying the
would sustain the more rapid growth re- organism with a sufficiency of the metals
quired for mass culture. Modification of cannot, however, be solved simply by add-
the inorganic medium, however, by ad- ing them to the medium. Evidence has
justing concentration and pH and adding been obtained by several investigators that
silicon in the case of the diatoms, was not certain micronutrients become biologically
sufficient. unavailable shortly after their incorpora-

The yield data indicate that although tion into the nutrient solution. This phe-
numerous species may be grown readily in nomenon may explain the progressive re-
large-scale culture under natural illumina- duction in yield from continuous mass cul-
tion, their productivities will differ. More tures of algae which demonstrate a high
rapid growth from any one species could yield only during the early period of
be expected only after intensive investiga- growth.

tion of nutrient requirements. Genetic bar- Numerous techniques have been devised
riers to rapid cell division undoubtedly in an effort to prevent the displacement of
exist in many species, making their ex- the trace metals, especially iron, from solu-
ploitation even in ideal media technically tions by co-precipitation. Among these the
difficult. Analyses of the samples now un- nonmetabolizable chelating agent ethylene-
der way, however, may indicate organic diamine tetraacetic acid (EDTA) is the
constituents of sufficient value to warrant most promising. Its chelating or complex-
further study of certain species. ing property prevents precipitation, but

permits sufficient release of minute amounts

Chelated Micronutrient Levels for of the micronutrients for plant metab-

SCENEDESMUS 4 olism.

Robert W Krauss ^he use °^ EDTA is attended with some

A , , , -it difficulties. Potassium does not form a

A problem commonly met in the culture , , . , -^t^^ A 1 1

f 1 1 r 1 • 1 1 11 • chelate with EDI A, and the magnesium

or algae and or higher plants as well is 1 1 • 11 1 1 • 1 i-

, ,? . 1 1 • c ano - calcium chelates have high dissocia-

the limitation on growth resulting from . ~, . ,

, r . t, , .,. , . , . tion constants. I his means that macronu-

dehciency or unavailability or essential in- . .„ , , , • 1 1

'. , , trients will probably not compete with the
organic micronutrients. With the excep- . . £ , ., , , t-t^t- a

. 1-11 if micronutrients tor the available ED 1 A.

tion or boron, which has not been clearlv n-u j- r l •

, . ; lhe dissociation constants tor the micro-

shown to be essential for algae, these micro- . • . 1 .1 1 1 1

. . r . nutrients, however, though much lower,

nutrients are all metals. Those for which

4 This work is being done at the University of
Maryland. by the cells, and in turn released by mass

are not identical, so that as they are ab-
4 This work is being done at the University of sorbed at different rates from the medium

i68

CARNEGIE INSTITUTION OF WASHINGTON

action from the complexes, competition of
residual EDTA for the remaining metals
will occur. The dissociation constants,
listed in the order of increasing degree of
stability, for the complexes at 20 ° C in
0.1 N KC1 are as follows: magnesium,
2.04 Xio -9 ; calcium, 2.56 Xio -11 ; manga-
nese, 3.40 Xio~ 15 ; iron (ferrous), 6.02 X
io~ 15 ; cobalt, 7.95x1c -17 ; zinc, 2.63 Xio -17 ;
copper, 4.17 Xio -19 ; and iron (ferric), iX
io~ 25 . A greater quantitative requirement
for one element, such as iron, will release a
progressively larger amount of EDTA,
which must of necessity reduce the avail-
ability of the remaining metals. This con-
sideration is further complicated by lack of
information concerning the equilibria at-
tained when a number of different ions
are in competition for EDTA at various
pH values or in the presence of different
buffers. Ideally, if the known dissociation
constants did hold for multimetallic solu-
tions, the most desirable chelating agent
would be one in which the order of dis-
sociation paralleled the order of quantita-
tive ion absorption. Sufficient information
concerning quantitative absorption and
multi-ionic equilibria does not yet exist,
however, to permit calculation of a precise
formula for supplying micronutrients as
chelates to algae. Until this can be done it
appears advisable to determine experimen-
tally the levels of chelated micronutrients
which give maximal growth.

The multiple culture apparatus previ-
ously described (Year Book No. 51, 1951-
1952), modified to provide a maximal ir-
radiance of 400 foot-candles, from 4500°
white fluorescent tubes, has been employed
in this study. The culture vessels have
been redesigned and are now equipped
with ground-glass joints, allowing for
greater protection against contamination.
Macronutrients were supplied as potas-
sium nitrate, magnesium sulfate, and

potassium-dihydrogen phosphate. Before
autoclaving, the micronutrients were in-
troduced individually from stock solutions
of reagent-grade inner-complex salts of
EDTA. An inoculum of Scenedesmus ob-
liquus, washed three times in double-dis-
tilled water, gave an initial packed cell
volume of 5 X io~ 5 ml/100 ml to each cul-
ture. The temperature was maintained at
25° C. The cultures were supplied with
5 per cent C0 2 in air at the rate of 0.5
liter per hour per culture of 200 ml.

TABLE 4

Effect of various concentrations of the
inner-complex salts of calcium, manga-
nese, cobalt, zinc, copper, and iron with
ethylenediamine tetraacetic acid on the
growth of Scenedesmus obliquus in the
presence of a complete macronutrient
solution

Concentration

Packed cell

of total metals

of each metal

volume per 100 ml

(ppm)

after 10 days

0.065

1.275

1.225

0.975

162

trace

486

0.000

With EDTA, a maximum increase in
total growth of 25 per cent was measured
as compared with cultures which were
supplied micronutrients in the form of
chlorides, nitrates, and sulfates. This in-
crease was observed both in the multiple
culture apparatus and in mass cultures.
Levels up to 500 ppm of each metal were
tested. Growth ceased at and above 100
ppm. The results of experiments, with
concentrations under 100 ppm, are given
in table 4.

The optimal range for growth was be-
tween 1 and 9 ppm for each micronutrient
or between 6 and 54 ppm of all the micro-
nutrients. It is likely that the inhibition of

DEPARTMENT OF PLANT BIOLOGY

169

growth observed at the higher concentra-
tions is due to a toxic effect of the metal
complement and not to the effect of one,
such as copper, which might be toxic per
se. This hypothesis has been tested by
experiments in which the concentration of
the copper chelate was varied, with each of
the other chelates held at 1 ppm. The re-
sults indicate that the optimum level for
copper is 3 ppm but that growth at higher
concentrations is impaired very little, be-
ing reduced by only 31 per cent at a copper

concentration of 81 ppm. Experiments to
determine the optimal and inhibitory levels
of each of the other micronutrients are
under way. These data should make pos-
sible a micronutrient formula of wide ap-
plicability in the culture of algae. In con-
junction with the production of large
quantities of algae in mass culture, it is
hoped that this technique may provide
quantitative data on micronutrient require-
ments and absorption under conditions of
maximal growth.

EXPERIMENTAL TAXONOMY

During the current year efforts have in new plantings at Stanford and at the
been directed toward the completion of the mountain transplant stations of the Insti-
Poa program and the development of new tution. Several new hybrid strains of
programs. Numerous conferences were promise were discovered, adding to the
held between members of the staff and number previously obtained. For sources
governmental and private agencies con- of seed of the many new Poa strains we are
cerning the most effective means of de- almost completely dependent on the Nurs-
veloping the new hybrid Poa lines al- ery Division of the Soil Conservation Serv-
ready produced and of familiarizing plant ice, which, during the past year, has
breeders with techniques of producing new carried on an active program of new plant-
lines of their own. From these discussions ings for seed increase of the hybrid Poas.

a co-operative research program has been
developed with the Department of Agron-
omy of Purdue University, Lafayette, Indi-
ana, and furthermore, a ten-state testing
program of the new Poa lines will be un-
dertaken by the Division of Forage Crops
and Diseases of the U. S. Bureau of Plant

Hybrids produced during a second pe-
riod of crossing in 1951 became mature in
the Stanford garden this year, and many
were cloned for testing at the altitudinal
transplant stations of the Institution. The
data from these hybrids are needed for
completing the basic biological studies on

Industry. Other developments include the Poa, and it is anticipated that other agen-

initiation of screening tests of new Poa cies may undertake further development of

hybrids by the Department of Agronomy any new strains that may have agronomic

of the University of California at Davis, interest.

California, and of some strains for turf and A detailed analysis of the complex ge-
range uses by the Ferry-Morse Seed Com- netic systems controlling the morphological
pany at Mountain View, California. expression of characters in contrasting
Hybrid lines developed from the first climatic races of the diploid species Poten-
series of crossings, performed during the tilla gland ulosa in three environments was
period between 1943 and 1945, were further concluded, and the results have been as-
studied in the extensive plantings at the sembled in manuscript form. It was found
Nursery Division of the Pacific Region of that the genetic systems of ecologic races
the U. S. Soil Conservation Service, and are even more complex than preliminary

170

CARNEGIE INSTITUTION OF WASHINGTON

analysis had indicated. The characters of
these climatic races are regulated by series
of multiple genes of different strength and
of opposing action, and of complementary
and activating genes, and of genes that ex-
press themselves only in certain environ-
ments. Only by a study of cloned indi-
viduals of the F 2 hybrid generation in
three contrasting environments over a pe-
riod of years combined with an analysis of
segregation in F3 populations was it pos-
sible to interpret the apparently conflicting
lines of evidence. This is the first time
that a genetic analysis has been carried out
on such a scale in more than one environ-
ment, and the results emphasize gaps in
our knowledge concerning the expression
of gene influences in different environ-
ments.

Another approach to the study of the
genetic structure of ecologic races is being
developed in Mimulus. Two diploid spe-
cies, M. car din alts from low altitudes and
M. Lewisii from alpine environments, were
intercrossed, and many Fi hybrids have
been obtained. These will be used as a
starting point for genetic analyses of these
morphologically and ecologically contrast-
ing racial complexes, utilizing the trans-
plant stations. It is also planned to use
these plants in connection with a compara-
tive physiological study of these contrast-
ing altitudinal races and their segregating
progeny.

Similar studies are being made on hy-
brids between contrasting climatic races of
Achillea lanulosa which are tetraploid
and which therefore probably have more
complex genetic systems. During the year
a series of Fi hybrids between lowland and
subalpine or alpine races of A. lanulosa
has been produced for this purpose.

Substantial progress in the construction
of an apparatus for comparative studies of
individual physiological processes in plants

over a range of controlled environmental
conditions has been made this year. It is
hoped that this equipment will lend itself
to comparative studies on contrasting eco-
logical races and on hybrid progenies as a
step in the analysis of the relation between
physiological differences and gene differ-
ences, especially as regards characters of
significance in natural selection.

Starting in September 1952, Grun began
a year's study at the laboratory of Professor
T. O. Caspersson at the Karolinska Insti-
tute in Stockholm, Sweden, in order to
study the chemical composition of cell con-
tents by the methods developed by Dr.
Caspersson. Ultraviolet microspectropho-
tometric techniques are used to determine
the nucleic acid and protein contents of
cells in both a dividing and a nondividing
state. The purpose is to gain a better un-
derstanding of the biochemical and phys-
ical changes taking place during cell di-
vision.

The Sixth International Grasslands Con-
gress, held at Pennsylvania State College
August 17 to 23, 1952, was attended by
Clausen and Grun, at which time Clausen
presented a paper on "New bluegrasses by
combining and rearranging genomes of
contrasting Poa species." Clausen and
Hiesey also attended the Western Grass
Breeders' Conference at Mandan, North
Dakota, June 22 to 24, 1953, where Clausen
presented a paper describing the Poa pro-
gram of the Institution. These conferences
offered opportunities to discuss the grass-
breeding investigations with agronomists
from different regions within the United
States and also from other countries.

The Poa Program

Jens Clausen, William M. Hiesey, and
Malcolm A. Nobs

A review of the progress of the range-
grass breeding program begun in 1943 was

DEPARTMENT OF PLANT BIOLOGY

171

included in the annual report of last year, and the Carnegie Institution, and have em-

During the current year, thought and phasized the interdependence involved in

effort have been directed toward conclud- this co-operation. The present stage of

ing this program, with the objective of development seems to warrant an exten-

making available to the public the gains sion of co-operation also to other agencies,

made both in basic research and in poten- especially state experiment stations and the

tially valuable new hybrid strains of blue- U. S. Bureau of Plant Industry, whose

grasses, and at the same time of conserv- scope of activity includes the breeding and

ing the efforts of the staff so that their time selection of forage plants,
may be devoted more completely to funda- Such extended collaboration is aimed at

mental biological investigations. The con- stimulating interest in the practical po-

traction of this program can most eflfec- tentialities of breeding new apomictic blue-

tively be accomplished through gradual grasses through wide interspecific crossing,

steps whereby other agencies well qualified and also at making possible widespread re-

to deal with practical agronomic prob- gional tests under current agronomic prac-

lems may take over aspects of the work tices. It was natural to start in California,

carried by the Institution that fall natu- where many of the strains have been

rally within the scope of their activities. developed. Accordingly, through Dr. R.

The contributions of the Carnegie Insti- Merton Love, the Department of Agron-
tution in its grass-breeding program fall omy of the University of California has
into two classes: first are the discoveries established small plot tests at the Univer-
of basic biological importance that clarify sity Farm at Davis, California, using 30
our understanding of the evolutionary pat- new hybrid lines and parent strains of Poa
terns and forces operating in apomictic developed by the Institution in co-opera-
genera, and also furnish essential informa- tion with the Soil Conservation Service,
tion to the plant breeder; second are the Another co-operative program has been
new hybrid strains of Poa that have been established with members of the Depart-
stabilized through apomixis and that may ment of Agronomy at Purdue University,
have economic value either in their present Dr. A. L. Peterson, head of the Depart-
state of development or when combined ment of Agronomy at Purdue, and Dr.
in still untried new crossing combinations. H. H. Kramer, cytogeneticist in the same

The large backlog of information which Department, visited the Poa plantings at

has been accumulated and is still growing the Stanford and Mather stations as well

must be analyzed and published in sum- as the Soil Conservation Service nursery

marized form in order that the basic at Pleasanton, California, during May of

knowledge that has been gained may be the current year. They plan a sustained

available both to scientists and to practical research program to explore the possibili-

plant breeders. This obligation in itself ties of developing new strains particularly

will require considerable time and effort, suited to meet their requirements in Indi-

The problem of making the best possible ana and adjacent regions. Steps have been

use of the strains and new hybrids of Poa taken to establish a fellowship supported

already developed has been given detailed jointly by the Carnegie Institution of

consideration, and steps have been taken in Washington and the Purdue Research

this direction. Previous Year Book reports Foundation for the special training of

have described the close co-operation be- qualified young agronomists in the tech-

tween the U. S. Soil Conservation Service niques of developing new strains through

172

CARNEGIE INSTITUTION OF WASHINGTON

wide interspecific crossing of apomictic Poa scabrella X P. pratensis originating

lines. This program is being started with from a segregating, sexual F 2 progeny,

a detailed screening test of the available These scabrella-pratensis F 2 derivatives

apomictic lines of Poa for possible direct were among the few scabrella-pratensis

use in Indiana or for further crossing. survivors at Pullman after a mild January

A comprehensive regional screening of followed by a cold February without snow
the more promising Poa strains that have that eliminated whole blocks of strains
been stabilized through apomixis is being previously thought to be winter-hardy,
developed by Dr. A. A. Hanson, of the The new lines are much more winter-
Office of Forage Crops and Diseases of the hardy and vigorous than are derivatives of
U. S. Bureau of Plant Industry. In this scabrella-pratensis of this same combina-
test some 30 of the more promising Poa tion previously grown at Pullman, and
lines will be planted in replicated test their morphological characters suggest that
plots at some ten co-operating experiment they arose through fertilization of a sexual
stations in the United States. The actual Fi hybrid by pollen of P. pratensis alpigena
planting of these tests will not be under originally from Lapland, which was grow-
way until the fall and spring of 1954-1955 ing in an adjacent plot at Pullman. The
because the necessary seed will not be selection of these new stabilized strains at
available until then. During the current Pullman in a combination formerly con-
year fairly extensive new plantings of sta- sidered to be wholly unsuccessful in this
bilized hybrid lines have been made by environment is a spectacular demonstration
the Soil Conservation Service at their nurs- of the possibility for successful selection
eries at Pullman and Pleasanton, from among open-pollinated progenies of hybrid
which seed may be available for such tests combinations that initially appear to be
as are being projected by the U. S. Bureau unpromising,
of Plant Industry. Especially gratifying is the stabilization

New data regarding the performance of an apomictic line of Poa ampla X P.
of existing strains have been obtained at compressa this year at Pullman after re-
the Soil Conservation Service nurseries at peated failures in previous years. This
San Fernando and Pleasanton in Cali- demonstrates that it is possible to combine
fornia, and at Bellingham and Pullman in the genomes of these contrasting species
Washington, in addition to new informa- into new apomictic lines once the proper
tion from the plantings at the Stanford, genie balance is achieved. The new hybrid
Mather, and Timberline transplant stations is vigorous at Pullman and has promising
maintained by the Carnegie Institution, characteristics, but further testing will be
Several new apomictic lines of Poa ampla required before its possible agronomic
X P. pratensis have been stabilized at the value can be determined. Another new
Pullman nursery of the Soil Conservation stabilized apomict was obtained from the
Service. These and the ones obtained in same F 3 progeny which evidently is the
1952 are F 2 lines that are definitely superior product of an open-pollinated outcross be-
to earlier lines of the same interspecific tween an F 2 hybrid of P. ampla X P. corn-
cross, being more leafy and more vigorous pressa and an adjacent culture of ampla-
in growth. These new types include some alpigena. The new derivative is a vigorous
of our most promising new hybrid lines. strain that combines characteristics of the

Among the new apomictic lines obtained three species, P. ampla, P. compressa, and

at Pullman are two constant derivatives of P. pratensis alpigena.

DEPARTMENT OF PLANT BIOLOGY

173

The improvement in quality of new
apomictic lines of hybrid Poas throughout
the years is clearly evident at the Pullman
nursery of the Soil Conservation Service,
where the 1948, 1949, 1950, 1951, and 1952
plantings could be compared directly dur-
ing the current year. The extensive data
that have been recorded in successive years
from the same cultures make possible a
fairly accurate evaluation of the merits of
these strains in the Pullman environment.
It is evident that only a small fraction of
the possibilities for selection among the

Cytology of Poa Hybrids

Edward L. Triplett

A basic step in the analysis of the cytol-
ogy of Poa hybrids is the determination of
their chromosome numbers, and most of
the effort during the current year has been
devoted to such a study of the hybrids re-
sulting from the 1951 crossings. The plants
which have been investigated fall into three
groups: (1) hybrids involving Poa arida,

(2) hybrids involving P. arachnifera, and

(3) quadruple hybrids involving three or
four species.

TABLE 5

Somatic chromosome numbers in parents and interspecific F x hybrids of Poa

Culture no.

Combination

Parents, 2w =

Fi hybrid individuals, 2n —

6298

P. arida X ampla

6299.... P. arida X ampla

6271 ... . P. scabrella X arachnifera

6272. ... P. arachnifera X pratensis

63, 56 95, 94, 92, 92, 92, 91, 91, 90, 90, 89, 89,

88, 86, 85, 77, 64
63, 56 99, 96, 92, 91, 90, 90, 90

84, 56 75, 75, 74, 73, 72, 70, 69

56, 68 65, 64, 63, 61, 61, 61, 61, ca. 54

possible hybrids in Poa have been realized
during the ten-year period since the Poa
breeding program was begun, and that it
should be possible to produce apomictic
strains of Poa having almost any combina-
tion of characteristics found among the
component species of the genus.

Current studies on the new Poa hybrids
reported in Year Book No. 51 (1951-1952),
pages 111-117, that are the result of the
1951 crossings indicate that many of the
new combinations are of evolutionary as
well as of possible agronomic interest, but
the data are still too incomplete to warrant
a detailed description of them. Cloned
individuals of these hybrids and their par-
ents are now being established at the three
transplant stations, and they have also been
the subject of studies on chromosome num-
bers reported below by Triplett.

Poa arida-ampla combinations. Two
plants of Poa arida, both originally from
North Platte, Nebraska, and having 2/2 = 63
chromosomes, were pollinated by an indi-
vidual of P. ampla from Albion, Washing-
ton, having 2/2=56 chromosomes. The
hybrids are intermediate between the par-
ents in morphology and are vigorous when
grown at Stanford. The chromosome
numbers indicate that 21 of a sample of
23 Fi plants of this combination were the
result of the union of an unreduced female
gamete and a reduced male gamete, as
shown in table 5. Two individuals devi-
ated from the usual range of 2/2 = 88 to
92 chromosomes in having only 64 and
77, and probably were the product of
haploid or hyperhaploid gametes from
both parents.

Poa arachnifera combinations. A plant

174

CARNEGIE INSTITUTION OF WASHINGTON

of Poa scabrella originally from Watson-
ville, California was pollinated by P. arach-
nifera (272 = 56) from Stillwater, Okla-
homa. The chromosome number of the
scabrella parent is yet undetermined but
was probably 84, the normal number for
this species. Seven of the Fi hybrids
ranged in chromosome number from 272 =
69 to 75 (table 5) . These hybrids are sum-
mer-active at Stanford, have short rhi-
zomes, and combine the characteristics of
the parent species.

hybrids segregate widely. Evolutionary
they are of considerable interest, and some
appear to have highly desirable agronomic
characteristics. Their chromosome num-
bers in connection with their morphology
provide some basis for deductions regard-
ing the possible genome composition of
these new derivatives. The chromosome
numbers of the original grandparental spe-
cies, the apomictic hybrid parent lines, and
a sample of their quadruple hybrid off-
spring are listed in table 6. The largest

TABLE 6

Somatic chromosome numbers in quadruple hybrid combinations of Poa

Culture
no.

Combination

Grandparents,

2w =

Fi parents,
2w =

Quadruple hybrid
individuals,

2n =

6301. ... P. arida-ampla X ampla-alpigena 63, 63, 70, 74 94, 73

6302. ... P. ampla-alpigena X arida-ampla 70, 74, 63, 63 73, 94

6303. ... P. arida-ampla X ampla-pratensis 63, 63, 63, 68 94, 95

6310. .. . P. scabrella-pratensis X ampla-alpigena 84, 68, 70, 74 75, 73

99, 85, 69, 65, 64

95,91

87, 83, 80

82, 80, 75, 73, 73,
72,71,71,71,71,
71, 69, 69, 68, 66,
66, 64, 64, 63, 63,
63, 62, 56

The same plant of P. arachnifera was
also pollinated by a plant of P. pratensis
from Mather, California, having 272 = 68
chromosomes. Eight Fi plants, all morpho-
logically indistinguishable from P. arach-
nifera, were vigorous at Stanford. One had
the chromosome number 272 = 54 anQl was
probably not a hybrid, but the other seven
ranged from 272 = 61 to 65, as shown in
table 5, and are apparently the product of
essentially haploid ovules combined with
haploid pollen.

Quadruple hybrids. The quadruple com-
binations studied included the following:

(1) Poa arida-ampla X ampla-alpigena,

(2) Poa ampla-alpigena X arida-ampla,

(3) Poa arida-ampla X ampla-pratensis,
and (4) Poa scabrella-pratensis X ampla-
alpigena. Progeny from these quadruple

sample is of 6310, a combination of ampla-
alpigena and scabrella-pratensis. It is ob-
vious from the sample of 23 plants that the
hybrids were produced from gametes hav-
ing reduced chromosome numbers. The
chromosomes in this remarkable quad-
ruple hybrid combine four very contrasting
species from highly contrasting environ-
ments, but range only between 272 = 56 and
272 = 82 with frequency peaks at 272 = 62-64
and 71-73.

Genetic Studies on Ecological Races

Malcolm A. Nobs, Jens Clausen, and
William M. Hiesey

Studies directed toward extending our
knowledge of the genetic structure of eco-
logical races belonging to the same or

DEPARTMENT OF PLANT BIOLOGY x ^

closely related species are being pursued segregate widely in the F 2 . Although

on a broad and systematic basis. They are morphologically very distinct, these two

directed along two major lines, aimed at species behave biologically as ecological

the analysis of the genetic structure of races of one species. Some transplants of

climatic races (i) on a diploid and (2) parents and hybrids were brought from

on a polyploid chromosome level. Stanford to Mather in April 1953, during

On the diploid level the genetic differ- a period of frost and snow, and were found
ences are basic, and although they are far to differ greatly in their degree of frost
from simple in diploid races of wild plants, resistance; cardinalis was highly susceptible
they are less complex than in polyploids, to frost injury, but the hybrids were far
Studies on diploids, therefore, hold some more resistant. Differences in rate of
promise of ultimate success in a genetic growth of seedling plants at different tem-
analysis designed to reveal the actions and peratures were also observed between the
interactions of specific genes. The studies parents and the hybrids: cardinalis seed-
on alpine and lowland races of Potentilla lings were favored in growth by warm
glandulosa mentioned in earlier Year days, Lewisii by cool days, and the Fi
Books fall into this category. The yet un- hybrids by both. These characteristics,
published results of the genetic analysis of coupled with an array of striking mor-
these races reveal the action of intricate phological differences among the hybrids,
systems of balances governed by series of make this group of plants particularly in-
genes that determine typical character dif- teresting for study of the mode of inherit-
ferences between the races. The races of ance of morphological and physiological
P. glandulosa have many features useful differences between ecologically distinct
for genetic analysis, but they are slow in races. One special use of these plants being
growth and development. contemplated is in connection with the

A highly promising group of hybrids physiological studies on climatic races men-
having most of the advantages of Poten- tioned in later paragraphs of this report. In
tilla glandulosa and being faster in develop- addition to crosses in the Mimulus cardi-
ment has been found in the Mimulus cardi- nalis-Lewisii complex, parallel crossings
nalis-Lewisii complex. The first hybrids are being tried between contrasting races
between M. cardinalis, a vigorous, red- and species in the M. guttatus complex,
flowered perennial herb of low altitudes, which, from Vickery's earlier work, were
and M. Lewisii, a pink-flowered herbaceous also shown to have important possibilities
perennial of high altitudes, were made by for this type of basic investigation.
Dr. Robert Vickery in the course of his On the polyploid level the genes in mul-
graduate research (cf. Year Book No. 50, tiple series also tend to be multiplied. The
1950-1951, p. 119). Additional hybrids, complex mode of inheritance found in
using strains differing in greater degree diploid Potentilla glandulosa suggests that
than those used in Vickery's original actual gene analysis in wild polyploid races
crosses, were made last year and during would be nearly impossible. Nevertheless,
the summer of 1953. These are being self- although not so clear-cut as studies on the
pollinated and backcrossed for further ge- diploid level, studies of crossings between
netic analysis. Both parents and hybrids polyploid races of the same or closely re-
have n = 8 pairs of chromosomes, and from lated species are essential to the under-
Vickery's experiments these are known to standing of evolutionary processes because

176

CARNEGIE INSTITUTION OF WASHINGTON

of the frequency of polyploids in both wild
and cultivated species. The various poly-
ploid members of the Achillea millefolium
complex have been extremely useful from
this point of view. In Year Book No. 51
(1951-1952), pages 122-124, some of these
hybrids and their progeny, especially be-
tween members on the hexaploid level, in-
cluding A. borealis, were described. Dur-
ing the current year a number of new Fi
combinations have been obtained, mostly
on the tetraploid level between different
races of A. lanulosa. These hybrids are
summarized as follows:

Port Chicago, salt marsh, coastal X Timber-
line, Sierran subalpine, 10,000 ft altitude

Port Chicago X Port Orford, Oregon coast

Dyerville, Redwood region, 250 ft altitude X
Timberline

Dyerville X Big Horn Lake, Sierran alpine,
11,700 ft altitude

Dyerville X Hoosier Ridge, Rocky Mt. al-
pine, 11,700 ft altitude

Port Orford X Timberline

Port Orford X Big Horn Lake

Port Orford X Hoosier Ridge

Of special interest is the hybrid between
the tall coastal form of A. lanulosa (earlier
referred to A. puberula Rydbg.) from the
salt marshes in the vicinity of Port Chicago,
Contra Costa County, California, and the
dwarf subalpine race of A. lanulosa ssp.
alpicola from near Timberline station at
10,000 feet altitude. This hybrid is being
self-pollinated to obtain an F 2 population.
If the progeny from this combination are
vigorous and show the segregation ex-
pected, it is planned to utilize the F2 in a
selection experiment at the Stanford,
Mather, and Timberline stations for com-
parison with a similar experiment already
under way using F 2 progeny of a hybrid
between contrasting latitudinal forms of
A. borealis from Kiska Island in the Aleu-
tian chain and from the San Joaquin Val-

ley in California, as described in Year Book
No. 51.

The new hybrids made this year between
races of A. lanulosa are of additional in-
terest in relation to the cytological irregu-
larities in the lanulosa hybrids, Port Or-
ford X Big Horn Lake and Port Orford
X Vera Cruz, reported by Ehrendorfer
last year (Year Book No. 51, pp. 124-125).
These irregularities suggest that genetic
barriers may exist between some races of
this highly polymorphic tetraploid species.
The new hybrids should make possible an
adequate study of this problem.

To judge from morphological observa-
tions, it appears that interspecific hybrids
between races of the hexaploid A. borealis
and the tetraploid species A. lanulosa made
by Ehrendorfer last year have also been
successful, but cytological and genetic con-
firmation has not yet been obtained.

Comparative Physiology of Climatic
Races

William M. Hiesey and Harold W. Milner

Natural selection is probably the most
important force governing the evolution
of plants. The studies on the genetic struc-
ture of climatic races of Potentilla reveal
that both morphological and physiological
differences are determined by complex
systems of gene balances. Used in this
sense, the term "physiological differences"
refers primarily to observed contrasts in
rates of growth, seasonal periodicity, frost
resistance, and survival when clones of in-
dividuals of each race are exposed to dif-
ferent environments. Observations on such
responses are of great value in interpreting
general relations between plants and their
environment, but they do not supply the
information needed to understand the spe-
cific mechanisms that operate in natural
selection. The role of differences in phys-
iological activity as a factor in natural se-

DEPARTMENT OF PLANT BIOLOGY

177

lection is not at all clear. Critical studies
on the variations in the individual physio-
logical processes in response to controlled
variations in environment are needed be-
fore we can begin to understand the rela-
tions between physiological processes and
natural selection.

Present methods for the measurement of
such basic physiological processes as photo-
synthesis and respiration cannot be applied
directly to the problem at hand. The ex-
isting methods are neither rapid enough
nor sufficiently flexible to permit precise
quantitative observation of the response of
many individual plants to a wide range of
controlled variation in each of several en-
vironmental factors. Variability within lo-
cal populations, races, and species also
needs to be taken into account. This neces-
sitates the study of many individuals, each
cultivated as a clone, under a wide range
of controlled environmental factors.

The work of the past year has been con-
cerned primarily with the development of
apparatus with these requirements in mind.
As was mentioned in Year Book No. 51
(1951-1952), pages 131-132, emphasis is
being placed on the study of rates of res-
piration and photosynthesis under con-
trolled conditions of temperature, light in-
tensity, and atmospheric composition, keep-
ing other external factors constant. Other
factors may be varied later as they appear
to be important.

The apparatus under construction is de-
signed to determine the rate of consump-
tion or liberation of carbon dioxide and of
oxygen by the plant in a recirculating air
stream in a closed system. Changes in the
concentration of carbon dioxide and of
oxygen in the air stream are determined by
analyzers which, in turn, control pumps
for injecting or absorbing the proper
amounts of the respective gases to maintain
their concentrations within predetermined

limits. The amount of carbon dioxide or
oxygen injected or withdrawn is recorded
automatically. The time lag between
changes in concentration of carbon dioxide
or oxygen in the plant chamber and their
adjustment by the automatic controls is
estimated at around 2 minutes. The ac-
curacy of control of the carbon dioxide or
oxygen concentration in the air stream
of the apparatus is expected to be approxi-
mately 1 per cent of the total concentra-
tion of either gas.

A sample of the circulating air stream
in the apparatus is continuously by-passed
first through the carbon dioxide analyzer
and then through the oxygen analyzer.
The carbon dioxide analyzer is of the type
designed by Dr. Moyer D. Thomas, of the
American Smelting and Refining Com-
pany. Dr. Thomas and his associate, Mr.
James O. Ivie, have been most generous in
supplying detailed sketches, photographs,
and verbal information that have been
invaluable in the construction of the ana-
lyzer. In the Thomas method the carbon
dioxide in the air sample is absorbed quan-
titatively by a standardized solution of
sodium hydroxide. The conductivity of the
resulting solution is related quantitatively
to the carbon dioxide concentration in the
air sample. Part of the carbon dioxide-
free air from the analyzer is then passed
through a Model F-3 Beckman Oxygen
Analyzer, which measures the concentra-
tion of oxygen by the paramagnetic
method developed by Dr. Linus Pauling
at the California Institute of Technology.
The concentrations of carbon dioxide and
oxygen are measured simultaneously and
continuously.

The electrical signals from the carbon
dioxide and oxygen analyzers go to sep-
arate amplifiers and electronic control de-
vices developed by Hawk. The electronic
controls have three functions: first, to ac-

T yS CARNEGIE INSTITUTION OF WASHINGTON

tivate the carbon dioxide and oxygen in- plants, their rapid growth, and their ready
jectors and absorbers; second, to provide increase through clonal division are ad-
signals to an Esterline-Angus multiple-pen vantages from the standpoint of practical
on-off recorder; third, to permit spot- experimentation. Their occurrence in
check readings on a voltmeter of the con- many parts of the world and in different
centration of either carbon dioxide or kinds of climate indicates that the Lem-
oxygen. naceae may contain a number of distinct
Zenith metering pumps, with their on ecological races and species. The biology
and off times recorded, accurately meas- of the species of this family, however, is
ure the volume of gas passing through the not sufficiently well known, and some pre-
injectors and absorbers. Injectors for car- liminary investigations are needed to deter-
bon dioxide and oxygen introduce the re- mine the suitability of members of this
quired amounts of these gases directly in- group for physiological studies of the kind
to the apparatus. Excess carbon dioxide is contemplated.

absorbed by passing a portion of the air Kruger has made valuable suggestions
through a solution of sodium hydroxide, in the mechanical design of the apparatus,
Excess oxygen is removed by injecting and his help in its construction is grate-
hydrogen into a sample of the air, which fully acknowledged,
then passes through a Deoxo Gas Purifier,

where the hydrogen reacts catalytically Ultraviolet Absorption Spectra of Iso-

with the stoichiometric equivalent of oxy- lated Root-Tip Nuclei

gen. The temperature and relative humid- p AUL Grun

itv of the recirculated air are regulated by ~ , , ,. t

3 , , .. ., . . , One approach to an understanding or

passage through a cooling coil maintained , " c „ ,. . . , . . x "

r ° ° ttt r the causes or cell division (mitosis) is to

at the proper temperature. Water or tran- , , , • i j-n- i

. , r . , ., . . . study the chemical differences between

spiration condenses in the coil and is col- n « ,. . , , , , ,

r cells that divide and those that do not.

lected and measured. A/r u i • n t i i- i • r n

Morphologically, metabolic nuclei or all

The completion and testing of the ap- cdls appear tQ be y£ry similar? although
paratus will require several additional some may be about tQ enter the complex
months. It will be necessary to adapt the stages of divisiori) and others may never
apparatus to the study of plants differing divide again> During a year spent in St0 ck-
in size and structure. Plant materials suit- holm> Sweden, at the laboratory of Profes-
able for such studies are available in abun- sor x. O. Caspersson, an attempt is being
dance from the concurrent investigations made to analyze the nucleic acid and pro-
being conducted in the Department. A tein contents of nuclei of cells still going
search for other plants even better suited through the mitotic cycle, and of cells in
to studies of the comparative physiology differentiated, nondividing tissue. This
of ecologic races is nevertheless being should offer the possibility of comparing
made. Of interest in this connection are nuclei that have just gone through divi-
members of the family Lemnaceae, and sion, those about to go through division,
some exploratory collections of Lemna and and those that would ordinarily never
Spirodela have already been made. These again divide.

are among the smallest of the higher plants Growing root tips of Tradescantia palu-

and have a relatively simple anatomical dosa contain dividing cells in large num-

structure. The very small size of these bers in a region restricted to within 2.5 mm

DEPARTMENT OF PLANT BIOLOGY

179

from the tip, whereas no divisions occur
in the following 2.5 mm. The part of the
root with nondividing cells is derived
from the mitotic portion with which it is
to be compared, and so contains cells that
ceased division in the recent past. A tech-
nique has been developed for freeing nu-
clei from cell walls and cytoplasm so that
they can be subjected to microscopic study
in an isolated condition. Roots are freeze-
dried, a process of killing and drying
thought to leave tissue in a relatively un-
changed condition with respect to chemical
content. They are then immersed directly
in dry glycerine and remain there for at
least 6 hours. Thereafter root parts still
in the glycerine are mashed and ground
with a spear-tipped needle, a process which
frees many intact nuclei, which float from
the cell walls and cytoplasm. The mixture
is then mounted on a quartz slide for
study.

The ultraviolet microspectrophotometric
methods developed by Professor Caspers-
son are being used to assay the chemical
content of the isolated nuclei. This method
involves moving individual nuclei across a
beam of ultraviolet light using a single
wave length at a time, and measuring elec-
tronically the fraction of light of each wave
length absorbed by the nuclei. The nuclei
are scanned at wave lengths of 2650, 2750,
2800, 2900, and 3150 A. It is known that
nucleic acids (both ribo- and desoxyribo-
nucleic acid) have an absorption curve
with a maximum at 2600 A, whereas the
only two amino acids of importance in
this connection, tryptophane and tyrosine,
have absorption maxima at about 2800 A.
The relative heights of the 2650 and 2800

A absorptions are, thus, a measure of the
relative content of nucleic acid and of
protein; and the absolute heights are a
measure of the total quantities of these
absorbing substances present.

The measured nuclear absorption curves
usually show a peak at 2650 A, and a hump
at 2800 A. There is great variation even
within single tissues in the sizes of the
nuclei, the heights of the absorption curves,
and the ratio of 2650 to 2800 A absorption.
To judge the amounts of nucleic acid and
amino acid present in the nuclei, it is neces-
sary to separate the complex absorption
curves into component parts. The assump-
tion is made at the start that the proportion
of tyrosine to tryptophane in the protein
is the same as that commonly found in
chemical analyses of plant proteins, 2.7
parts tyrosine to 1.6 parts tryptophane.
Curves of the absorption of mixtures of
different proportions of nucleic acid and
the amino acids have been calculated to
fit as closely as possible the observed nu-
clear absorption. It has been necessary
also to take into account the fact that the
position and height of the tyrosine maxi-
mum varies with the degree of dissocia-
tion in the molecule. Good fits have been
obtained between the observed nuclear ab-
sorption spectra and calculated nucleic
acid-amino acid absorption spectra when
calculations were based on 25 per cent dis-
sociation of the tyrosine.

Work is now in progress on amassing
a large enough body of material so that
the statistical significance of the differences
between nuclei of these tissues can be
determined.

i8o

CARNEGIE INSTITUTION OF WASHINGTON

PALEOBOTANY

Ralph W. Chaney

During the 1952 field season, plant fos-
sils were collected at two localities in Ore-
gon where the mixed deciduous forest of
the Oligocene epoch is well preserved.
One of these is at a recently discovered site
in the John Day Basin, in beds assigned
on stratigraphic grounds to the Lower
John Day formation. The second fossil
flora comes from Rujada in the Willamette
Valley on the western side of the Cascades;
stratigraphic relations are here less clear,
for the plant-bearing sediments are inter-
bedded in lava flows which may span a
large segment of the Tertiary period. A
comparison of the two floras has been
made by Dr. R. W. Lakhanpal, of the
Birbal Sahni Memorial Institute of Paleo-
botany, in the course of a year spent as
UNESCO fellow at the University of
California. They are similar in their pre-
dominance of temperate genera, such as
Alnus (alder), Quercus (oak), and Plata-
nus (sycamore), and have many identical
species in common. But the Rujada flora
has a more conspicuous element of broad-
leafed evergreens, which appear to have
made up an important part of the under-
story. Another significant difference is the
rarity of Metasequoia (Chinese redwood)
at Rujada and the occurrence there of
Sequoia of the coast redwood type; by
contrast, Sequoia has not been collected at
Fossil, where the foliage of Metasequoia
outnumbers the remains of all the asso-
ciated trees. It seems apparent that the
Rujada forest was a coastal assemblage, in
which marine climate and its lower tem-
perature range made possible the occur-
rence of many evergreen plants; by con-
trast, the Fossil forest, living some hundred
and fifty miles from the Tertiary shore,
was largely deciduous and more typically
temperate. In any such interpretation the

present-day restriction of Sequoia semper-
virens to the coastal belt of California and
Oregon, and the contrasting occurrence of
Metasequoia glyptostroboides in the in-
terior valleys of central China, provide the
basis for an understanding of Oligocene
forests in which their Tertiary ancestors,
Sequoia affinis and Metasequoia occiden-
tals, were dominant trees. Today, with
the more recently uplifted Cascade Range
intervening to form a major climatic bar-
rier, the forests of the Willamette Valley
and the John Day Basin show even greater
differences than during the Oligocene.

Stratigraphic paleontologists have long
looked for similarities rather than differ-
ences in determining the age of deposits
bearing plant and animal fossils. In recent
years students of fossil invertebrates, es-
pecially microfossils, have begun to recog-
nize the significance of facies differences,
and paleoecology now enters largely into
their considerations of age. Vertebrate pa-
leontologists have been somewhat slower
to recognize the possibility that two unlike
faunas may have a close age relationship,
though even in this field of life history
there is evidence of increasing understand-
ing of the areal relations of Tertiary mam-
mals. The combination of similarities and
dissimilarities in the case of the Rujada
and Fossil floras, and its bearing on the
problem of the age of the deposits con-
taining them, is one of many such cases
which the study of Tertiary floras in
western North America has brought to
light. Having in mind the movements of
vegetation in response to climatic change
in later geologic time, it may be said that
complete similarity of two fossil floras
from widely separated localities is a chal-
lenge to any conclusion that they lived

DEPARTMENT OF PLANT BIOLOGY

181

at the same time. In the past, as today,
there have been widespread contempora-
neous forests; but any careful analysis of
the details of their composition is likely to
show the same sorts of differences, result-
ing from varying latitude, altitude, and
distance from the sea, that may be noted
in comparisons of modern vegetation.

Using this dynamic approach to the
problems of age determination, compari-
sons are now being made between the
well known floras of the standard section
in the John Day Basin and recently dis-
covered floras in western Oregon and
Washington. The active assistance of Mr.
Rudolf Erickson and Mrs. Eleanor Gor-
don, of Portland, both of whom have

discovered and collected significant plant-
bearing deposits, along with other mem-
bers of the Geological Society of the
Oregon Country, is adding greatly to the
effectiveness of the study of Tertiary vege-
tation west of the Cascades. In north-
western Washington Marie B. Pabst has
under way a comprehensive program deal-
ing with the older Tertiary floras there,
long known but little understood. Her
detailed study of the ferns of the Chucka-
nut formation is already adding significant
evidence regarding the climate and age of
early Tertiary floras long grouped together
under the Puget group, and is showing
that the latter name no longer has either
stratigraphic or floristic significance.

BIBLIOGRAPHY

Chaney, Ralph W. Conifer dominants in the
middle Tertiary of the John Day Basin, Ore-
gon. The Paleobotanist, vol. i, pp. 105-113
(1952).

Davis, Edwin A. Photosynthetic Chlorella mu-
tants. Amer. Jour. Bot., vol. 39, pp. 535-539
(1952).

Jean Dedrick, C. S. French, H. W.

Milner, Jack Myers, J. H. C. Smith, and
H. A. Spoehr. Laboratory experiments on
Chlorella culture at the Carnegie Institution
of Washington Department of Plant Biology.
In Carnegie Inst. Wash. Pub. 600, 1 pp. 105-

153-
Dedrick, Jean. See Davis, Edwin A.

French, C. S. See Davis, Edwin A.

Grant, Verne. Cytogenetics of the hybrid Gilia
millejoliata X achilleaejolia. I. Variations in
meiosis and polyploidy rate as affected by
nutritional and genetic conditions. Chromo-
soma, vol. 5, pp. 372-390 095 2 )-

Isolation and hybridization between

Aquilegia jormosa and A. pubescens. El
Aliso, vol. 2, pp. 341-360 (1952).

Genetic and taxonomic studies in Gilia.

II. Gilia capitata abrotanijolia. El Aliso,
vol. 2, pp. 361-373 (1952).

Genetic and taxonomic studies in Gilia.

III. The Gilia tricolor complex. El Aliso,
vol. 2, pp. 375-388 (1952).

Hase, E. See Tamiya, H.

Hiesey, William M. Growth and development
of species and hybrids of Poa under con-
trolled conditions. Amer. Jour. Bot., vol. 40,
pp. 205-221 (1953).

Dynamisme de revolution d'apres la

conception d'un experimentateur. In Col-
loque International du Centre National de
la Recherche Scientifique sur l'Evolution et
la Phylogenie chez les Vegetaux, vol. 5, pp.
C271-C279 (1952).

See Juhren, Marcella.

Iwamura, T. See Tamiya, H.

Juhren, Marcella, William M. Hiesey, and

F. W. Went. Germination and early growth

of grasses in controlled conditions. Ecology,

vol. 34, pp. 288-300 (1953).
Kok, Bessel. Experiments on photosynthesis by

Chlorella in flashing light. In Carnegie Inst.

Wash. Pub. 600, 1 pp. 63-75.
Krauss, Robert W. Inorganic nutrition of algae.

In Carnegie Inst. Wash. Pub. 600, 1 pp. 85-

102.
and William J. McAleer. Growth and

evaluation of species of algae with regard

to sterol content. In Carnegie Inst. Wash.

Pub. 600, 1 pp. 316-325.

1 Carnegie Institution of Washington Publication
600. Algal culture: From laboratory to pilot plant.
Editor, John S. Burlew. ix + 357 pp. (1953).

l82

CARNEGIE INSTITUTION OF WASHINGTON

Macdowall, F. D. H. The reducing potential
of illuminated chloroplasts. Science, vol. 116,

PP- 398-399 (1952).
MacGinitie, Harry D. Fossil plants of the

Florissant beds, Colorado. (Contributions to

Paleontology.) Carnegie Inst. Wash. Pub.

599. iii + 198 pp. (1953).
McAleer, William J. See Krauss, Robert W.
Milner, Harold W. Chlorella as a source of

proteins and fats: Laboratory to pilot plant

experiments. The Vortex, vol. 13, pp. 368—

380 (1952).

The chemical composition of algae. In

Carnegie Inst. Wash. Pub. 600, 1 pp. 285-302.

See Davis, Edwin A.

Mituya, A., T. Nyunoya, and H. Tamiya. Pre-
pilot-plant experiments on algal mass culture.
In Carnegie Inst. Wash. Pub. 600, 1 pp. 273-
281.

See Tamiya, H.

Myers, Jack. The biology of the algae: A brief
summary. In Carnegie Inst. Wash. Pub. 600, 1
PP- 31-36.

. Growth characteristics of algae in rela-
tion to the problems of mass culture. In
Carnegie Inst. Wash. Pub. 600, 1 pp. 37-54.

See Davis, Edwin A.; Sorokin, Con-

STANTINE.

Nihei, T. See Tamiya, H.
Nyunoya, T. See Mituya, A.
Sasa, T. See Tamiya, H.

Shibata, K. See Tamiya, H.

Smith, J. H. C. See Davis, Edwin A.

Sorokin, Constantine, and Jack Myers. A
high-temperature strain of Chlorella. Sci-
ence, vol. 117, pp. 330-331 (1953).

Spoehr, H. A. English summary of "StofTpro-
duktion durch Griinalgen und Diatomeen in
Massenkultur," by H. von Witsch and R.
Harder. In Carnegie Inst. Wash. Pub. 600, 1
pp. 164-165.

The need for a new source of food. In

Carnegie Inst. Wash. Pub. 600, 1 pp. 24-28.

Society in the grip of science. Address

of retiring president, Pacific Division,
A. A. A. S. Published by the author. 16 pp.
(1952).

A new look at the food problem. Gar-
den Jour. New York Bot. Garden, vol. 2, pp.
133, 134, 153, 156 (1952).

See Davis, Edwin A.

Tamiya, H., E. Hase, K. Shibata, A. Mituya,
T. Iwamura, T. Nihei, and T. Sasa. Kinetics
of growth of Chlorella, with special refer-
ence to its dependence on quantity of avail-
able light and on temperature. In Carnegie
Inst. Wash. Pub. 600, 1 pp. 204-232.

See Mituya, A.

Went, F. W. See Juhren, Marcella.

1 Carnegie Institution of Washington Publication
600. Algal culture: From laboratory to pilot plant.
Editor, John S. Burlew. ix + 357 PP- (i953)-

DEPARTMENT OF EMBRYOLOGY

Baltimore, Maryland
GEORGE W. CORNER, Director

During the past year the staff of the De- ence will increase the competence of the
partment of Embryology, augmented by Department to deal with the biochemical
one major appointment, has maintained its problems, now pressing for solution, of the
program of research and publication at full living egg and embryo and the female re-
tide. The only serious problem has been productive system.

to find room for all the workers and for Dr. Beni Horvath, formerly Fellow of

the numerous visitors who come from all Columbia University, College of Physi-

parts of the world to avail themselves of cians and Surgeons, joined the Department

the collections and experimental facilities. July i, 1952, on a fellowship of the Car-

Dr. David W. Bishop became a member negie Institution, as a member of the
of the Department October 1, 1952. A group working with Dr. Csapo on the bio-
graduate of Swarthmore College and of chemistry of uterine muscle,
the University of Pennsylvania (Ph.D. Dr. Brenda M. Schofield came in Janu-
1942), he was with the U. S. Fish and ary 1953 from the Department of Pharma-
Wildlife Service (1941-1942) and the cology of Oxford University, England, on
Army Air Force Physiology Program a Fulbright Fellowship.
(1942-1946). After holding academic posi- Dr. Ronald Singer completed in Septem-
tions at the University of Colorado and the Der 1952 his term as Fellow of the Interna-
University of Illinois, he became professor t i ona i R ota ry Foundation, and returned to
of physiology at the University of Massa- n i s post at the University of Cape Town,
chusetts (1948-1951). In 1951-1952 he held Soutn Africa.

a fellowship of the Planned Parenthood Dr Vittorio Danesino, Department of

Foundation, working at the California In- Gynecology and Obstetrics, University of

stitute of Technology. His researches have Naples, began in April 1953 a stay of a

dealt chiefly with the physiology and bio- year as a Fulbright Fellow, during which

chemistry of sperm cells, with special refer- ne wi n_ wor k i n association with Dr. Bar-

ence to oxygen utilization and enzymatic telmez, Dr. Bishop, and Dr. Reynolds, and

activities. The methods used in this kind a l so i n tne Department of Obstetrics, Johns

of research will ultimately be useful also Hopkins Hospital.

in analyzing similar vital activities of the Dr. Chester H. Heuser, Research Asso-

early embryo and of its uterine environ- c i ate) aga i n S p en t several weeks at the labo-

ment. Like the sperm cells, the fertilized ratory, in the summer of 1952, continuing

eggs and early embryos of mammals live his work on very early embryos,

for a time free in the reproductive tract, Dr. E. Carl Sensenig returned for ten

depending on their internal resources to weeks in the summer of 1952 to continue

utilize oxygen and nutritive substances fur- investigations on the development of the

nished by the surrounding tissues of the spinal column in human embryos. At the

mother. It is expected that as Dr. Bishop University of Alabama he has prepared

follows out his well developed program of about twenty human fetuses put at his dis-

research on the male germ cells, his experi- posal on loan, for study of the later stages

183

i8 4

CARNEGIE INSTITUTION OF WASHINGTON

of normal and abnormal development of Dr. R. Licata, University of Michigan,

the spine. July-August 1952: development of coro-

Commander Roy E. Crowder, Medical nary vessels.

Corps, U. S. Navy, again devoted two Dr. Eileen Otis, University of Rochester,

weeks of his leave (December 1952) to re- October-November 1952: variability of rate

search on the development of the human of development of human embryos,

adrenal gland. Dr. Bradley M. Patten, University of

Dr. Perry W. Gilbert, of Cornell Uni- Michigan, February 1953: study of human

versity, made two visits of a few days each embryos.

to work on the illustrations of his forth- Dr. C. A. Salvatore, University of Sao

coming monograph on the development of Paulo, Brazil, October 1952: to observe the

the external muscles of the human eye. program of investigations.

The following persons also visited the Dr. Luis Vargas, Catholic University of

laboratory, making use of the collection of Santiago, Chile, March 1953: to observe the

human and other primate embryos, for program of investigations,

study of the topics indicated: Absence of Director. The Director was

Dr. James K. Avery, University of Roch- abroad most of the year, having been

ester, January 1953: development of the gran ted leave of absence from September

teeth. 1, 1952 in order that he might accept an in-

Dr. Elio Borghese, University of Pavia, vitation to serve as George Eastman Visit-
Italy, September-November 1952: early ing Professor at the University of Oxford
human embryos. for the academic year 1952-1953.

Dr. Gordon W. Douglas, New York When informing Dr. Corner of this ac-

University, May 1953: development of tion of the Trustees, Dr. Bush emphasized

mammalian embryos to trophoblast stage, their earnest wish to contribute in every

Dr. Kenneth B. Duke, Duke University, possible way to Anglo-American co-opera-
June 1953 : study of primate embryos in the tion in science. At Oxford Dr. Corner
Bluntschli Collection. received a warm welcome and generous

Dr. A. M. Elshout, New York Univer- courtesy at all times from the University

sity, May-June 1953 : modeling techniques, officials, from the Department of Anatomy,

Dr. H. Goldberg, Johns Hopkins Hospi- and from Balliol College, to which he was

tal : frozen-section technique. attached as Fellow for the year.

Dr. Hugh G. Grady, Armed Forces In- Dr. Samuel R. M. Reynolds served as

stitute of Pathology, July 1952: human Acting Director during this period,

trophoblast and decidua. Academic honors. Dr. Robert K. Burns

Dr. Robert C. Hardy, New York Hospi- received the honorary degree of Doctor of

tal, July 1953 : study of embryos with hare- Science from his alma mater, Bridgewater

lip. College, June 1, 1953.

Dr. Miriam Kelner, University of Re- Dr. Arthur T. Hertig, Research Associ-

cife, Brazil, December 1952: study of hu- ate, was appointed Shattuck Professor of

man embryos. Pathological Anatomy and Head of the

Dr. E. Levy, Morrisania Hospital, New Department of Pathology, Harvard Uni-

York, April-August 1953: development of versity, beginning with the autumn term

pelvic floor and anal region. of 1952.

DEPARTMENT OF EMBRYOLOGY

185

Lewis H. Weed

The death of Dr. Lewis H. Weed in De-
cember 1952 is appropriately recorded by
the President elsewhere in this Year Book.
The Department of Embryology, however,
has special reason to mourn his loss, for in
various ways he was associated with it for
thirty-five years. One of his earliest scien-
tific papers — in the long run perhaps the
most valuable of all his contributions — the
monograph on The Development of the
Cerebro-Spinal Spaces in Pig and Man,
beautifully illustrated by James F. Didusch,
was published in 1917 as volume 5 of the
Contributions to Embryology. Two sub-
sequent articles appeared in volumes 9
(1920) and 17 (1926). From 1922 to 1935
Dr. Weed, who was during those years
professor of anatomy in the Johns Hopkins
University, was also a Research Associate
of the Carnegie Institution, receiving an
annual grant for technical assistance in his
valuable researches on the cerebrospinal
fluid and related subjects. As a Trustee of
the Institution from 1935 until his death,
he was in a favorable position, in the most
literal sense, to keep an eye on the Depart-
ment of Embryology from his own office
next door, in the Johns Hopkins Depart-
ment of Anatomy. Dr. Weed was associ-
ated with the founder and first director of
the Department of Embryology, Professor
Franklin P. Mall, as a student and later as
a member of Mall's staff in the Medical
School. He was the closest scientific neigh-
bor and friend of the second director, Dr.
George L. Streeter, during the whole of his
term of office. To the present incumbent,
particularly during the earlier years of his
service, before duties at the National Re-
search Council began to keep Dr. Weed at
Washington most of the time, the latter
was particularly helpful, combining per-
sonal friendship with ever wise and dis-
creet advice.

Contributions to Embryology,
Volume 35

At the time of this report, volume 35 of
the Contributions to Embryology is in
preparation. Publication is expected in
1954. The volume will contain the follow-
ing articles:

231. George W. Corner and George W.
Bartelmez, Early abnormal embryos of the
rhesus mon\ey.

232. Christine Gilbert and Chester H.
Heuser, Studies in the development of the
baboon {Papio ursinus). A description of
two presomite and two late somite stage em-
bryos.

233. G. W. Bartelmez, with the collabora-
tion of Mary P. Blount, The formation of
neural crest from the primary optic vesicle in
man.

234. S. R. M. Reynolds, G. M. Ardran, and
M. M. L. Prichard, Observations on regional
circulation times in the lamb under fetal and
neonatal conditions.

235. L. J. Wells and G. van Wagenen,
Androgen-induced female pseudohermaphro-
ditism in the mon\ey (Macaca mulatto):
Anatomy of the reproductive organs.

236. L. J. Wells, Development of the hu-
man diaphragm and pleural sacs.

237. Anna W. Chacko and S. R. M. Reyn-
olds, Architecture of distended and nondis-
tended human umbilical cord tissues, with
special reference to the arteries and veins.

238. Elizabeth Mapelsden Ramsey, Venous
drainage of the placenta of the rhesus mon\ey
{Macaca mulatto).

239. Theodore W. Torrey, The early de-
velopment of the human nephros.

240. Arthur T. Hertig, John Rock, Eleanor
C. Adams, and William J. Mulligan, On the
preim plantation stages of the human ovum:
A description of four normal and four ab-
normal specimens ranging from the second
to the fifth day of development.

241. George W. Settle, Localization of
the erythrocyte-forming areas in the early
chic\ blastoderm cultivated in vitro.

i86

CARNEGIE INSTITUTION OF WASHINGTON

RESEARCH IN PROGRESS

Embryo, Fetus, Uterus

Early human embryos. Dr. Chester H.
Heuser, Research Associate, continued his
work on the earliest embryos of the Car-
negie Collection, that is to say those of
groups i to v of the Streeter classification.
He has made reconstructions of several
early embryos of groups vi and viii. Dr.
Heuser plans to prepare monographs of
these stages which will form a companion
volume to that of Dr. Streeter (see Year
Book No. 51, 1951-1952, p. 168), this being
the period of development that was not yet
covered at the time of Dr. Streeter's death.
Dr. Heuser has added three excellent em-
bryos of horizon viii (about 17 days old)
to the collection. Two of these (nos. 9009A,
B) are identical twins, by far the earliest
human specimens of twinning yet seen.
The case is especially valuable because it
gives indisputable proof that uniovular
twinning can actually occur in one of the
ways that have been postulated from the
evidence of later stages, namely by forma-
tion, inside a single blastocyst, of two em-
bryonic areas, each of which becomes a
separate embryo.

Dr. Arthur T. Hertig, Research Associ-
ate, reports that during the year five hu-
man uteri, thought likely to contain very
early embryos, were searched without suc-
cess. Several excellent embryos of later
stages were sent to Baltimore for the de-
partmental collection. Dr. Hertig and his
associates in Boston are working on a
micro-biochemical study of enzymes, inor-
ganic iron, and calcium in human embryos.
This work, supported by the Carnegie In-
stitution only in so far as the Institution's
grant on behalf of the search for early em-
bryos facilitates also the collection of later
stages which can be utilized for the chemi-
cal study, is of course closely relevant to
the Department's general program. An in-

teresting detail of the findings, some of
which are in press at the time of this re-
port, is that the primary germ cells (cf . the
account of these cells by Dr. Emil Witschi
in Contrib. to Embryol., vol. 32, pp. 67-80)
exhibit a high alkaline phosphatase activity
which sharply contrasts them with the sur-
rounding tissues.

Extrinsic muscles of the eye. Dr. Perry
W. Gilbert, of Cornell University, whose
frequent visits to the laboratory following
a long stay some years ago have been men-
tioned in these reports, has made consider-
able progress on his monograph on the ori-
gin and development of the human extrin-
sic eye muscles. Illustrations are now being
prepared by James F. Didusch. This is ex-
pected to be a definitive work, presenting
for the first time an exhaustive account of
a very important subject about which prac-
tically nothing appears in the literature of
human embryology. Dr. Gilbert's unex-
pected demonstration of "head cavities" in
human embryos (Year Book No. 50, 1950-
1951, p. 143) calls for renewed attention to
lower mammals. Head cavities have been
seen in a number of marsupials, but oddly
enough not in the North American opos-
sum (Didelphys virginiand) , which is one
of the most primitive of the marsupials.
Applying the experience gained from his
previous work to specimens belonging to
Dr. C. H. Heuser and to the Wistar Insti-
tute of Anatomy and Biology, Dr. Gilbert
has now found head cavities in opossum
embryos of the 10th to 12th days of gesta-
tion (McCrady's stages 28 to 32). As in
other vertebrates in which head cavities ap-
pear, they are obliterated by the primor-
dial eye-muscle tissue which develops about
them.

Effects of rubella {German measles).
As previously reported, the Collection now
contains several human fetuses (8 at the

DEPARTMENT OF EMBRYOLOGY

187

time of writing) that were aborted by phy-
sicians because the mothers had contracted
German measles during the first trimes-
ter of pregnancy. The fetuses in utero
therefore had become liable (according to
clinical experience) to severe injury to the
eyes, ears, or heart. Any information bear-
ing on the nature of this injury, on the
time in prenatal life at which it becomes
visible, and on the frequency with which
it is to be expected will be of great use to
physicians who have the grave responsi-
bility of advising a course of action when
a pregnant woman is affected by this other-
wise innocuous disease. Dr. G. W. Bartel-
mez is therefore continuing intensive study
of our specimens, all of which have been
serially sectioned. Thus far very little evi-
dence of damage has been found. One of
the specimens shows possible injury to the
lenses of the eye, but this must be checked
against a control specimen of the same age
prepared with the same fixative. Such a
specimen is not yet at hand.

Blood vessels of the endometrium. The
blood vessels of the lining of the uterus, in
man and rhesus monkey, are known to un-
dergo periodic structural changes related to
the menstrual cycle. Dr. Bartelmez is
making a detailed study of these changes,
using the extensive rhesus monkey mate-
rial available in the laboratory (Bartelmez-
Van Dyke, Hartman, Corner collections).
He finds that differences between speci-
mens of the same stage of the cycle are
more obvious in the vessels than in glands
and stroma, and believes that the distinc-
tive reactions of the four zones of the
endometrium are correlated with these
differences.

Umbilical cord. In connection with the
physiological studies of Dr. Reynolds on
the blood flow through the umbilical cord,
it became necessary to know something
about the nerve supply of the umbilical

blood vessels. Dr. Bartelmez has been ap-
plying his expert knowledge of neurology
to the elucidation of this problem. Dr. Vit-
torio Danesino is aiding in this study.

Nerves of the uterus. Dr. G. W. Corner,
while at Oxford, joined with Dr. Graham
Weddell, lecturer in anatomy there, and
Dr. Wazir Pallie, a postgraduate student,
in a study of the intrinsic nerves of the
uterus. It has long been known that the
motility of the organ is largely automatic,
and yet is influenced by factors that seem
to operate through the nervous system. In
experimental animals, stimulation of the
sympathetic nerves causes the uterine mus-
cle to contract. On the other hand, recent
investigations by Dr. Csapo and other
members of this Department have empha-
sized the independent contractility of uter-
ine muscles and have shown that such con-
trol as the body has over it is largely
exerted by the ovarian hormones. It there-
fore becomes necessary to learn as much
as possible about the nerve supply of the
uterus, in order to understand what part it
plays in uterine function. The study made
at Oxford, which utilized the most recent
improvements in intra-vitam staining of
nerves, developed by Dr. Weddell and his
associates, showed that the innervation of
the myometrium is of very simple, gener-
alized nature, resembling that of the blood
vessels. An illustrated report is being pre-
pared for publication.

Placenta. Thanks to the co-operation of
our obstetrical colleagues at the Johns Hop-
kins, Sinai, and Lutheran Hospitals, of
Baltimore, Dr. Elizabeth M. Ramsey now
has at her disposal a collection, remarkable
for extent and excellence, of 12 human
placentas in situ in the uterus, from all
three trimesters of pregnancy, in which
the fetal blood vessels have been injected.
Work on the analysis of the placental cir-
culation in these specimens is actively in

i88

CARNEGIE INSTITUTION OF WASHINGTON

progress, with particular reference to the
much-discussed Spanner theory of exclu-
sive marginal drainage. Dr. Ramsey's thor-
ough study of this question in the rhesus
monkey is now in press (Contributions to
Embryology) .

Pathology of the embryo and fetus. Dr.
Ramsey examined during the year 90 speci-
mens sent from eight states. Reports on all
these were sent to the physicians and insti-
tutions from which they came. Dr. G. W.
Bartelmez acted as consultant in some of
these cases. This routine task is consid-
ered worth the effort of a research labora-
tory, not only as a public service, but also
because it suggests problems and from
time to time yields valuable material for
study. Another activity of Dr. Ramsey in
this same field will receive mention below
under "Co-operative Research Activities."

Experimental Embryology

Recent progress made by Dr. R. K.
Burns in elucidating the effects of sex-
gland hormones on the embryonic urino-
genital system was fully discussed in Year
Book No. 51 (1951-1952), pages 170-171.
During the past year Dr. Burns has pur-
sued this important line of investigation
with particular reference to the effects
(discovered by him) of the sex-gland hor-
mones on the differentiation of the testis,
and to the reaction of the interstitial tissue
of the opossum testis to these hormones.

Physiology of implantation. As has pre-
viously been reported, Dr. Bent G. Boving
is investigating the mechanisms by which
the mammalian embryo becomes attached
and implanted in the uterus. He has found
that, in the rabbit, there are three phases,

(1) a muscular grasping by the uterus,

(2) an adhesion to the uterus, and (3) the
long-recognized implantation by groups of
embryonic trophoblast cells which invade
the uterus.

At the time of attachment, the products
of conception have developed into a blasto-
cyst, that is, a hollow sphere of cells with
a clump of cells, the future embryo proper,
at one pole. The blastocysts have enlarged
and have reached a stage at which they
are no longer moved longitudinally in the
cylindrical uterus, but are accommodated
by local ballooning out of the uterine wall
on the free side, opposite the uterine liga-
ment or mesometrium. This forms a
"dome" for each blastocyst. Dr. Boving
has conducted numerous experiments
aimed at understanding the mechanism
and function of dome formation. Sus-
pected mechanical and chemical stimuli
were imitated by distending the uterus
with plastic beads of various sizes, and
with similar beads the perforations of
which had been loaded with alkali. He
also measured the pressures needed to
make the uterus balloon out when dis-
tended by solutions of various degrees of
alkalinity. These experiments, which are
of preliminary nature, will be continued.
Dome function was studied by observa-
tions and manipulations of blastocysts
within the living uterus. Each blastocyst
was discovered to be held in its dome by
a muscular grasp which could be released
by injecting fluid until the uterus became
distended.

Similar procedure was found not to re-
lease blastocysts that were slightly more
advanced; they had become adherent to
the uterus. This attachment consistently
begins in a part of the blastocyst opposite
the embryonic rudiment, and it is accom-
panied by localized physical changes, in-
cluding the development of adhesiveness,
in the membranes surrounding the blasto-
cyst. Dr. Boving discovered that a corre-
sponding area shows a well defined alka-
line reaction in blastocysts removed from
the uterus and placed in appropriate test

DEPARTMENT OF EMBRYOLOGY

189

solutions. In the same region, by simply
drying a blastocyst on a glass slide, he
found crystals resembling sodium bicar-
bonate. The alkaline blastocyst substance
has a titration curve which resembles that
of sodium bicarbonate solution and shares
with it an increase of alkalinity on ex-
posure to air (presumably through a loss
of carbon dioxide) and a decrease of alka-
linity when expired air is blown through
the solution.

Dr. Boving has also given much thought
to the factors which determine the direc-
tion of growth of the trophoblastic proces-
ses as they sprout from the blastocyst wall
and invade the uterus during implantation.
His reconstruction and statistical analysis
of the pattern of growth indicates that in-
vasion is directed toward individual blood
vessels immediately underneath the uterine
epithelium. The guiding influence is al-
most certainly a chemical substance ema-
nating from the vessels. It is conceivable
that removal of carbon dioxide through
the vessels could produce an appropriately
localized alkalinity and adhesiveness of
trophoblast cells preceding actual invasion,
but the pattern of invading cell groups
suggests that the actual penetration of the
uterine epithelium covering the vessels
probably does not result from a direct ac-
tion on embryonic tissue by blood-borne
oxygen, acidity regulators, or enzymes, and
is presumably dependent on an indirect
action such as hormone-induced activity of
the uterine epithelium.

Fetal Physiology

Shortening of blood vessels. In a study
completed last year by Dr. S. R. M. Reyn-
olds and Dr. Anna W. Chacko but not
yet published, the architecture of umbilical
arteries and veins in the normally dis-
tended state was described. One of the
observations was that the degree of thick-

ening of the wall of an artery when empty
and constricted is such that it can be ex-
plained only by assuming concomitant
shortening and narrowing of the vessel.
The shortening was calculated to be of the
order of 17 per cent. This is made possible
by the double (clockwise and counter-
clockwise) spiraling of the muscular com-
ponents and the longitudinal arrangement
of elastic tissue in a distended, functioning
vessel. This year Dr. Reynolds was able
to measure the shortening, using lamb fe-
tuses delivered by Caesarean section. With
the cord pulsating, threads were tied onto
the wall of an artery at a known distance
apart. After constriction of the artery, the
distance was measured again. The amount
of shortening was thus ascertained to be
between 14 and 16 per cent. Similar meas-
urements on the ductus arteriosus showed
a shortening of 20 to nearly 40 per cent.
Thus shortening is an important compo-
nent of vasoconstriction.

Fetal vascular physiology. Twenty-five
fetal lambs that were available during the
winter were used by Dr. Reynolds prima-
rily to study the mechanism of a physio-
logical response of the fetal circulation,
that is, a sudden decrease in heart rate,
which is taken by obstetricians to indicate
fetal distress, such as occurs, for example,
when the umbilical cord is accidentally
compressed. The current explanation is
that an increase in fetal arterial pressure
due to blockage of the umbilical arteries
sets up reflex responses which oppose the
rise; these would include slowing of the
heartbeat. This hypothesis, for which there
is some support in the literature, is based
on a supposition that the fetus in utero is
reacting like an adult in the outside world.
This is not the case.

Dr. Reynolds finds that the heart rate is
slowed when the umbilical arteries alone
are occluded (with a rise in blood pres-

!Q CARNEGIE INSTITUTION OF WASHINGTON

sure) and also when the umbilical veins necessary. Equipment for the work is now

alone are occluded (with a fall in blood being installed.

pressure). The immediate slowing is abol- Amniotic fluid. Several years ago, Dr.
ished by cutting the nerve paths by which L. B. Flexner, then in this Department,
the heart rate is controlled. This shows and his associates established that in two
that nervous impulses arising from the species which they studied (guinea pig
venous channels leading from the placenta and man) amniotic fluid is continuously
to the heart are capable of regulating the formed and absorbed, with an increasing
action of the heart. It appears that the turnover of water and presumably there-
carotid sinus reflex path (via the 9th era- fore of the other constituents of the fluid,
nial nerve nucleus), the sympathetic sys- No one knows what are the sites of for-
tem (lateral reticulospinal tracts), and the mation, nor where the fluid leaves the am-
vagus nerves (10th cranial nucleus) are not niotic space in which it is enclosed. Dr.
functioning at a high level of activity, if at Reynolds has had an incidental opportu-
all, when the fetus is at rest. When, how- nity, during his observations on fetal cir-
ever, it is subjected to stress, these mecha- culation in lambs, to go into this question,
nisms are called into play. Figuratively In experiments in which the head of the
speaking, while the fetus is at rest the fetus was enclosed in a rubber bag, the
wires are in place and the connections latter became full of secretions from the
made, but the lines are not busy. With mouth cavity and nasopharynx. This oc-
stress, they go into action. curred even though the esophagus and

It was found that the resting fetal heart trachea were still open as possible avenues
rate is governed, in a broad sense, by fetal of exit, by swallowing or absorption re-
blood pressure. As the one increases, so spectively. As much as 30 cc was collected
does the other, and vice versa. in 45 minutes. This observation points to

When the fetal heart is completely freed at least one source of an appreciable por-

of its sympathetic and parasympathetic tion of the amniotic fluid, and accounts for

nerve connections, its beat becomes ex- the hitherto unexplained presence of a pro-

tremely rapid when the fetus is subjected teid substance, mucin, in this fluid,
to circulatory distress. The stimulus for

this increase is blood-borne; it is removed Biochemistry of Spermatozoa and of the
by adrenalectomy, and the active substance Genital Canals
is probably, therefore, adrenaline. These With the facilities of the Department of
considerations shed light on the puzzling Gynecology, Johns Hopkins Hospital and
clinical picture that is seen when, as some- Medical School, the role of uterine mo-
times happens in human labor, the fetal tility, including that of the cervix uteri,
heart is slowed (bradycardia) by circula- in sperm transport in the human female is
tory distress, and then later becomes exces- being investigated by Dr. D. W. Bishop,
sively fast. The latter state is probably an It appears that in women who are artifi-
indication that the infant is in extremis. daily inseminated just before complete

A fuller understanding of these prob- hysterectomy, no penetration of sperma-

lems requires study of the course and char- tozoa into the uterus occurs except dur-

acteristics of blood flow from the placenta ing the ovulatory phase of the cycle. This

through the fetal liver and into the heart; supports the findings of a number of

for such a study X-ray cinematography is workers who have stated that in experi-

DEPARTMENT OF EMBRYOLOGY

I 9 I

ments done in vitro, spermatozoa pene-
trate the cervical mucus only when the
latter is produced at mid-interval of the
menstrual cycle, and that live spermatozoa
can be aspirated from the uterus only if
insemination takes place at about the time
of ovulation. These observations are being
continued as cases become available at the
Hospital.

Experiments are under way to deter-
mine what role, if any, oxytocin plays in
the process of sperm transport through the
rabbit cervix. It has been suggested by re-
cent American and British investigators
that an oxytocic factor facilitates sperm
migration through the cervix of the cow,
and that oxytocin is released by copulation
in the rabbit. Direct evidence of this
mechanism is being sought.

Dr. Bishop's studies, begun elsewhere,
on the nature of the metabolic substrates
utilized by the spermatozoa during their
sojourn in the male genital ducts and dur-
ing their passage through the female tract
are being followed up by a chromato-
graphic study of the hexose sugars con-
tributed by the seminal vesicles and pros-
tate gland. Of particular importance is
the problem as to the degree of phos-
phorylation of the monosaccharides when
they are secreted. Two possibilities have
been suggested. Either free fructose or
glucose is secreted by the accessory glands;
or fructo-phosphate or gluco-phosphate is
secreted and subsequently dephosphoryl-
ated in the semen. If the latter alternative
is the correct one, significance is given to
the presence of very large quantities of
acid phosphatase supplied by the prostate
gland and, incidentally, also by the female
genital epithelium.

Hydrogen-ion concentration in the ovi-
ducts. A micro-method for determination
of hydrogen-ion concentration inside hol-
low organs, which Dr. Bishop previously
used in an investigation of the rabbit's vas

deferens, is being adapted to use in the
oviduct. Whereas the hydrogen-ion con-
centrations in the uterine cavity are fairly
well known, those of the oviducts (Fal-
lopian tubes) have not been determined.
These pH studies bear on three important
points: (1) the effect of pH changes on
the "fertilizin-antifertilizin" type of reac-
tion between gametes which Dr. Bishop
studied last year at the California Institute
of Technology; (2) the suggestion, made
by investigators elsewhere, that a high
tubal pH (8.5-9.0) may be responsible for
the dispersion of the cumulus cells around
the ovum; and (3) Dr. B. G. Boving's
discovery of a localized region of high
alkalinity in unimplanted rabbit blasto-
cysts.

Biochemistry and Physiology of
Uterine Muscle

The investigation of uterine muscle
physiology is going ahead so rapidly and
in such a sequential way that it is difficult,
when reporting its status, to avoid a con-
fusing overlap of results already published
with the statement of work in progress.
The reader is therefore advised to turn, at
this point, to the corresponding section
under "Published Research" in the latter
part of the report, and only after reading
that account to continue with the follow-
ing summary of current phases of the
program.

During the past year Dr. Csapo obtained
further evidence that intracellular potas-
sium is involved in the contraction pat-
tern of uterine muscle. This was given by
the observation that small doses of proges-
terone placed in the bath of physiological
salt solution in which a strip of estrogen-
dominated uterine muscle is contracting
will result in a slight increase in tension,
probably by lowering the intracellular po-
tassium from the superoptimal to the op-

192

CARNEGIE INSTITUTION OF WASHINGTON

timal concentration. If more progesterone
is put in the bath, intracellular potassium
will decrease below the optimal concen-
tration and submaximal tension will be
elicited.

Potassium-sodium ratio in uterine mus-
cle. However much the foregoing experi-
ments appear to support the working hy-
pothesis, they are not conclusive. Final
proof can be obtained by showing that the
intracellular potassium concentration, or
the ratio K:Na, in the progesterone-dom-
inated uterus is lower than that in the es-
trogen-dominated uterus. To demonstrate
this requires, as a preliminary, accurate
determination of the extracellular space
of the myometrium, about which nothing
is known. Dr. Beni Horvath, who under-
took this investigation, had to determine
extracellular space as well as potassium-
sodium concentration in rabbit uteri under
the domination of each of the two ovarian
hormones. His findings, which will be re-
ported in full in a journal article, indicate
(as was predicted from the physiological
experiments with living muscle) that un-
der progesterone domination the concen-
tration of intracellular potassium and the
K:Na ratio are lower than under estrogen.

Uterine muscle contraction in situ. If
our growing knowledge of the funda-
mental nature of uterine muscle contrac-
tion is to be useful in understanding the
behavior of the uterus as an organ, and
also those disturbances of uterine function
with which the obstetrician is confronted,
it will be necessary to find out whether the
living, intact uterus responds to electrical
stimulation in the same way as do isolated
strips of muscle studied in vitro. Dr.
Brenda M. Schofield joined the Depart-
ment temporarily in order to carry out
studies of this sort, for which she had be-
gun at St. Andrews University (Dundee,
Scotland) to work out a technical method.
In Baltimore the technique has now been

developed sufficiently to permit accurate
measurement of the isometric tension de-
veloped by the uterus in situ in the living
animal. It has thus been found that the
maximum work done by uterine muscle
in vivo and in vitro is similar, and that be-
havior in response to electrical stimulation
is also similar. Dr. Schofield's most in-
structive finding thus far is that the posi-
tive and negative "staircases" of estrogen-
and progesterone-dominated uteri occur in
the living animal, closely resembling, when
recorded graphically, the pattern previ-
ously observed in vitro by Dr. Csapo and
Dr. Corner.

Significance of measuring isometric ten-
sion. It has been the common practice, in
studying the physiology of the myome-
trium, to observe isotonic contraction by
observing the movements of a light lever
to which the muscle strip is attached. Dr.
Reynolds emphasized long ago that con-
clusions derived from such observations
are frequently contradictory and do not
give a coherent picture of the action of
hormones and oxytocic drugs. Dr. Csapo
finds that uterine muscle contracting iso-
tonically is insensitive (within certain lim-
its) to changes in temperature and in in-
tensity of electrical stimulation. On the
other hand, muscle contracting isomet-
rically (i.e., pulling against an excessive
load so that it does not, practically speak-
ing, shorten in doing its work) responds in
a strictly quantitative way to such changes
in temperature and stimulus. His tentative
explanation of these facts is that the uter-
ine muscle is composed of a number of
functional units. When under light load,
or no load at all, maximum shortening can
occur with only partial activation of these
functional units. On the other hand, under
optimum load the work done (or under
isometric conditions, the tension devel-
oped) quantitatively indicates the extent of

DEPARTMENT OF EMBRYOLOGY

193

activation, that is to say, the relative num-
ber of functioning units.

From these considerations it is clear that
the effect of drugs and hormones must be
observed isometrically (or isotonically but
with optimum load) if correct conclusions
are to be drawn from the recorded con-

tractions. Preliminary experiments with a
well known oxytocic hormone, pitocin, not
only illustrate this point, but give a new
concept of the nature of pitocin action. A
detailed account of these experiments and
the new views to which they lead is in
preparation for publication.

CO-OPERATIVE RESEARCH ACTIVITIES

Following upon previously reported co-
operation by Dr. Elizabeth M. Ramsey in
a study of prenatal mortality which is be-
ing conducted at the Columbia University-
Presbyterian Hospital Medical Center,
New York City, under the direction of Dr.
Katharine K. Merritt, Dr. Ramsey has ac-
cepted appointment as a member of the
advisory committee on a two-year continu-
ation study of gestation, birth, and the first
year of postnatal life by combined obstet-
rical, pediatric, and pathological investiga-
tion. Dr. Ramsey will continue to examine
fetuses sent from New York City in con-
nection with this investigation.

Joint study of a possible cause of human
infertility has been conducted by Dr.
Bishop in collaboration with Dr. Georgi-
ana Seegar-Jones, of the Department of
Gynecology of Johns Hopkins University.
The latter has isolated from several in-
fertile women a vaginal fungus which is
believed to have spermicidal properties.
Attempts to establish this fungus in the
vaginas of female monkeys have not yet
been successful.

On February 1, 1938 the Department of
Embryology provided headquarters in its
laboratory for a small group of scientific
men led by Dr. Robert K. Enders, working
on a study of reproduction in the mink
under the auspices of the U. S. Depart-

ment of Agriculture. The impetus for this
project had come in part from Dr. Carl G.
Hartman, then a member of the Depart-
ment of Embryology. After a few months
Dr. Enders was appointed to the faculty
of Swarthmore College and his project was
carried on there for three years, until it
was transferred to the Fish and Wildlife
Service, U. S. Department of the Interior.
In 1946 the work was transferred to the
Bureau of Animal Industry, U. S. Depart-
ment of Agriculture.

After fourteen years of intensive study
Dr. Enders has published a report which
constitutes a great advance in knowledge
of the breeding and reproduction of this
very important fur-bearing animal. In his
report he expresses gratitude to Dr. George
L. Streeter, Director of the Department
of Embryology at the time when the
Carnegie Institution fostered the project.
Credit is also given to Osborne O. Heard,
of the Department's staff, for technical
photography, and to Arthur G. Rever for
assistance in experimental surgical opera-
tions. The Department of Embryology is
pleased to see the evidence of a successful
investigation in a field so close to its own
interests, and to have had the opportunity
of encouraging this excellent work in its
earliest days.

THE CARNEGIE COLLECTION OF EMBRYOS

Through the efforts of Dr. G. W. Bar- University of Chicago has generously
telmez, the Department of Anatomy of the added to its previous gifts another group of

194

CARNEGIE INSTITUTION OF WASHINGTON

human embryos collected and prepared by
Dr. Bartelmez when he was at Chicago.
A special collection of serial sections of
human embryonic limbs was received from
the same source.

The Department of Embryology has also
acquired a collection of serial sections of
rat embryos in late somite stages prepared
by the late Professor F. L. Landacre, of
Ohio State University. With the addition
of these specimens to a set of ioo serially
sectioned rat embryos in early somite states,
in the possession of Dr. Bartelmez, the De-
partment is well equipped for study of
the development of the albino rat, an ani-
mal which is much used in experimental
embryology and for which, therefore, good
morphological material is necessary for
comparison and control.

Dr. Elizabeth M. Ramsey and Miss Har-
riet L. Caspari have made a most valuable
contribution to the administration and
housekeeping of the Collection by going
through all the stored specimens which
were shelved in bottles and jars, and cull-
ing out all items which were found to have
no research or museum value.

Richard D. Grill, photographer, and
Miss Caspari, with the advice of Dr. Bar-

telmez, have completed an arduous, time-
consuming review of the immense collec-
tion of photographic negatives, numbering
approximately 16,000 items. Useless and
redundant negatives have been discarded;
those of permanent value have been re-
checked. This valuable archive is in con-
stant use, not only to supply pictures
needed for research and publication by the
staff, but also to meet requests from embry-
ologists elsewhere. It has always been kept
in excellent order and is catalogued so that
desired negatives are instantly available.

Photographic records and models. Os-
borne O. Heard prepared complete serial
microphotographs of 12 presomite em-
bryos. These serial pictures greatly facili-
tate the process of modeling by the ad-
vanced techniques in use at the laboratory,
and are also useful for direct study, at least
of details not requiring examination of the
original sections, which are thus spared
much handling and are less frequently ex-
posed to the risk of breakage. Five older
embryos were also photographed in com-
plete serial sections. The remarkable early
human twin embryo mentioned above (no.
9009) was modeled by Mr. Heard.

PUBLISHED RESEARCH

Variability of the Menstrual Cycle

The study of nature by scientists involves
a constant effort to make an orderly ar-
rangement of the facts observed, which
often seem at first to be quite unrelated. In
biology the facts usually cannot be taken
out of their very complex setting and sub-
jected to detached consideration, in any
such way as that, for example, by which an
astronomer can theoretically consider two
heavenly bodies as geometrically regular
and affecting each other independently of
all other influences. Similar isolation of bio-
logical phenomena generally has to neg-

lect essential but imperfectly understood
considerations without which the most ten-
tative calculations are unsafe. Because, in
spite of this, the effort to discover basic re-
lationships in the life process must be
made, biological thinking tends, even more
than that of physical scientists, to alterna-
tion in its approach. The first step in ana-
lyzing a complex phenomenon is to work
out a general explanatory theory or codifi-
cation. This is accepted as a working basis
and often gains currency in diagrammatic
form. Then the pendulum swings. The
exceptions and contradictions to the ac-

DEPARTMENT OF EMBRYOLOGY

195

cepted "law" or principle begin to obtrude
themselves and are critically pitted against
the generalized theory. The diagram that
seemed to be adequate or at least useful
must often be discarded or redrawn.

A process of this inevitable sort has oc-
curred in the analysis of the female repro-
ductive cycle. By the efforts of numerous
biologists and physicians, beginning about
1890, the once mysterious recurrent phe-
nomenon of menstruation in the human
female, and the obviously related but
strangely diverse periodic sexual phenom-
ena in other mammals, have been ana-
lyzed. The way in which these cycles are
controlled by interaction of the ovarian
hormones with those of the pituitary gland
is understood in principle, even if not yet
completely in detail. The once baffling ex-
ternal differences between the human men-
strual cycle and the estrous cycles of the
familiar domestic animals have been ex-
plained, and a unified general concept of
the mammalian cycle has been established.

In all this work the natural tendency of
investigators has been toward a simplified
concept of the cycle. As a result, an over-
rigid picture has tended to develop, par-
ticularly with respect to the primate men-
strual cycle. For purposes of comparison
of one cycle with another, a convenient pat-
tern of the cycle has been constructed upon
the modal menstrual interval of 28 days.
Changes occurring in the uterus and else-
where have been conformed to the 28-day
schedule in an over-precise way, and the
variability of the intrinsic phases of the
cycle, for instance the preovulatory and
postovulatory periods, has been too much
neglected. The quantitative precision of
embryological data (such as the age of
very early human embryos), of physiologi-
cal hypotheses about cyclic events, and of
clinical diagnoses concerning abnormalities
of the cycle is in danger of overemphasis.

Dr. G. W. Bartelmez has presented, in

his study of factors in the variability of the
menstrual cycle, a cogent exposition of this
variability and of the limits which it im-
poses upon the precision with which inter-
nal events of the cycle can be timed from
clinical information, that is, the menstrual
history. His essay is the result of three dec-
ades of personal research on the cycle in
women and the rhesus monkey, as well as
of extensive acquaintance with the litera-
ture. This is not, fortunately, one of those
situations in which critical reconsideration
of the data threatens to render the whole
current theory obsolete; but it certainly
warns against overdiagraming of the work-
ing concepts. Dr. Bartelmez' summary of
his article follows.

Abrupt fluctuations in successive men-
strual cycles may occur even in "regular"
individuals, and errors are likely to result
from the application of statistical modes
and means to particular cycles. Mathemati-
cally all cycles in a statistical group are
treated as identical, yet two cycles with sim-
ilar external features such as length may
differ fundamentally even to the extent of
one's being ovulatory, the other anovula-
tory. Tests and other signs of ovulation
may be positive when the dominant follicle
fails to rupture and is resorbed or becomes
cystic, and the particular cycle is ano-
vulatory.

The fluctuations in the time relations
during the cycle appear to be due to varia-
tions in the ebb and flow of the controlling
hormonal tides. A long cycle may be due
to a long postmenstrual (quiescent) phase,
to a long-lived corpus luteum, or to a long
premenstrual (ischemic) phase.

The time of ovulation even in the rhesus
monkey has rarely been determined with
an accuracy of less than 12 hours, and the
time that may elapse between insemination
and fertilization is unknown in primates.
It is known that even normal siblings may
develop at different rates, so that one can-

196

CARNEGIE INSTITUTION OF WASHINGTON

not extrapolate with confidence from the
most accurately timed cases to others less
satisfactorily dated. Consequently we are
rarely in a position to state the age of a
pregnancy in any primate with greater ac-
curacy than ±2 days.

The morphological changes in the ova-
ries provide the only satisfactory criteria
for determining the sequence of events
during the cycle. The target organs are less
satisfactory, for they are subject to other
factors than hormonal tides. This is true
also of systematic changes such as fluctua-
tions in basal temperature. In abundant
material from rhesus monkeys, the endo-
metria have been classified on the basis of
ovarian findings, and obvious differences
have been found in endometria associated
with similar ovarian conditions. Even
when these endometria are arranged in as
many as eight groups corresponding to the
phases of the ovarian cycle, there are speci-
mens which exhibit some features of one
group and others of the succeeding group.

Umbilical Cord and Ductus Venosus

Sphincter of the ductus venosus. The
ductus venosus of the fetus is a direct chan-
nel from the umbilical vein to the inferior
vena cava, by which blood returning from
the placenta is enabled to by-pass the fetal
liver. In its wall there is a muscular
sphincter, only recently discovered, which
controls the rate of flow through the duc-
tus and thus aids in maintaining physio-
logical pressure in the umbilical vein. This
control must of course be necessary to the
fetus; and it also probably protects the um-
bilical cord against dangerous kinks by
keeping the cord vessels turgid. A knowl-
edge of the development of the sphincter
of the ductus venosus is obviously impor-
tant in understanding the pattern of um-
bilical blood flow in the growing embryo
and fetus. Dr. Anna W. Chacko and Dr.

Reynolds have published a study of this
question.

When the human embryo is 6.5 mm
long (crown-rump length), the ductus ve-
nosus is already well formed. At this time
the incipient sphincter first becomes visible
as a thickening of the endothelium at the
junction of the umbilical recess and ductus
venosus. As the embryo grows, the sphinc-
ter is more elaborately organized, and by
the 49th day of intrauterine life it has as-
sumed its definitive form and structure. It
contains circular and longitudinal fibers of
smooth muscle, intermingled with elastic
tissue. It is innervated by nerve fibers of
vagal origin, coming from the region of
the inferior vena cava and umbilical recess.

Connective tissue (Wharton's jelly) of
the umbilical cord. Pictured cross sections
of the umbilical cord, such as are presented
in textbooks, usually show the cord vessels
emptied of their blood. The cord thus
appears to be composed largely of loose
connective tissue of the embryonic type
(Wharton's jelly), in which run relatively
narrow arteries and veins, occupying only
a small part of the cross-sectional area. Dr.
Reynolds, in his article on this subject, pre-
sents photographs of cords of the human
species, sheep, and goat, which show the
comparison between cords in which the
blood vessels are full of blood as in life, and
others collapsed as they are when drained
of blood. In the functional condition the
expanded vessels occupy most of the cross-
sectional areas, crowding the Wharton's
jelly between the vessels and against the
capsular tissue of the cord, until the con-
nective tissue is hardly more than an ad-
ventitial sheath tautly drawn around the
arteries and veins. When the vessels con-
strict on being emptied, they shorten and
thicken, and the Wharton's jelly shortens
and thickens also. The occurrence of short-
ening of the vessels as an element in con-

DEPARTMENT OF EMBRYOLOGY

197

striction is being studied (see above under
"Research in Progress") and will be dis-
cussed in a forthcoming publication.

In the course of this study, Dr. Reynolds
was able to elucidate the nature of the
"folds of Hoboken" described in 1669 by
the Dutch anatomist of that name. These
are not valves as Hoboken thought, but
crescentic transverse bands of the lining
(intima) of the vessels. They presumably
exist in life, for they can be seen in umbili-
cal arteries which have been fixed in the
distended state without ever having been
allowed to collapse. Their function, if any,
remains obscure.

Blood flow in the umbilical arteries. In
Year Book No. 51 (1951-1952), a paper on
this subject by Dr. Reynolds and those
with whom he worked at the Nuffield
Institute, Oxford, in 1951-1952 was ab-
stracted. This included the statement, sur-
prising at first sight, that the umbilical ar-
teries do not dilate and contract with the
pulse wave. Their diameter remains con-
stant, within the limits of very close meas-
urement, throughout the cycle of pulsation.
Dr. Reynolds, with Dr. F. W. Light, of
Johns Hopkins University, Baltimore, and
Dr. Ardran and Dr. Prichard, of Oxford,
has now presented a full statement on the
qualitative nature of pulsatile flow in the
umbilical blood vessels. A summary of the
conclusions follows.

Insertion into an umbilical artery of a
needle for recording blood pressure un-
avoidably distorts the wall of the vessel
and the streamlines of flow. For this rea-
son true pressures like those in the intact
artery cannot be recorded by this method.
True lateral pressures were measured from
the first branch of an umbilical artery. The
vessel was fixed at right angles to the axis
of the main vessel, and the pressure trans-
mitted by means of a capacitance manome-
ter for recording.

When a long segment of umbilical cord

was directly photographed by X-ray mov-
ing pictures, using Thorotrast to make the
vessel visible, and recording pressures as
described above, it was found that at no
time in the pulse pressure wave was there
a demonstrable change in diameter of the
artery. Similarly, there was no difference
in the diameter of the artery along its
course such as would be expected if, owing
to frictional loss of energy, there were a
significant decrease of pressure within the
artery.

Pressure pulses were recorded in the um-
bilical artery and umbilical vein simultane-
ously. They were 180° out of phase with
each other, since the pulse pressure in the
vein decreased as that in the artery in-
creased, and vice versa.

Study of streamlines of blood flow by
X-ray cineangiography show that when a
quick velocity component of the moving
fluid during an arterial pulse wave hits
part of the umbilical artery, the vessel
may be displaced slightly. Such movement
gives rise to visible arterial pulsation.
There is no increase in arterial diameter
(except in the arch of the aorta) measur-
able by the methods used in this study,
that is, in direct cineangiograms recorded
at a frequency of twenty frames per second.

Normal umbilical vein pressure is high
if the flow is unobstructed, being equiva-
lent to 30-40 mm of mercury.

The rate of blood flow in the umbilical
vessels is very variable, being inversely re-
lated to the diameter of the vessels. The
volume of flow in the two arteries per kilo-
gram of body weight per minute is be-
tween 300 and 400 cc. This is about two-
thirds the volume that goes through the
descending aorta.

Fetal Circulation

In the fetus, blood flows from the pulmo-
nary artery to the aorta through the ductus
arteriosus. It is therefore reasonable to as-

198

CARNEGIE INSTITUTION OF WASHINGTON

sume that the pressure in the pulmonary
trunk is greater than that in the aorta. In
the adult animal, on the other hand, the
pulmonary arterial pressure is very much
less than the aortic pressure. Presumably
the change takes place at or near the time
of birth; but available technical methods
have been inadequate for exact timing of
this event and for the solution of many re-
lated problems of fetal blood circulation.
In a paper by Ardran, Dawes, Prichard,
and Wyatt of Oxford, and Reynolds of the
Department of Embryology, an account is
given of an investigation carried out at the
Nuffield Institute of Medical Research, at
the instigation of Dr. Reynolds.

By taking X-ray motion pictures at a
speed of 25 frames per second, cinemato-
grams of the fetal and neonatal circulation
were obtained which correctly represented
the cardiac cycle. It was found, contrary
to expectation, that before respiration be-
gins there is virtually no circulation of
blood through the lung of the fetal lamb.
Almost two-thirds of the blood put out by
the heart goes to the placenta. The earlier
study by Reynolds, Light, and co-workers,
cited above, shows that the factor which
controls the amount of blood going to the
placenta is the cardiac output, not the size
or length of the umbilical cord, nor any
limiting factor in the placenta itself. In
this mechanism it is essential that the di-
ameter of the umbilical arteries and veins
be ample, so that the energy lost by blood
flowing in the cord shall be very small in
comparison with the total energy of the
moving blood.

For the first time, in the work reported
here blood pressure was measured in the
fetal pulmonary trunk and aorta, and at
each end of the ductus arteriosus. In the
fetus, blood flows through the ductus un-
der considerable pressure. When the infant
begins to breathe, the blood is immediately
diverted from the ductus arteriosus into the

pulmonary arteries; and as the lungs ex-
pand, the volume rate of flow through
them increases almost fivefold. The pres-
sure in the pulmonary trunk drops to a
low point. Since the flow through the
lungs has increased concomitantly with a
fall in pressure, it must be inferred that
the resistance to flow has in some way been
reduced by the distention of the air spaces.
During the first few minutes after this ini-
tial change, the systemic blood pressure
also falls, apparently because of the trans-
fer to the lung of a significant portion of
the total blood of the infant. This transi-
tory general fall of blood pressure at the
onset of respiratory movements, which has
not previously been observed, may (if
overlarge) be dangerous. Several obstetri-
cians who have attended medical meetings
at which Dr. Reynolds discussed these find-
ings have already made use of them with
good effect in the management of circula-
tory conditions in the newborn infant.

Effective closure of the ductus arteriosus
was found to follow immediately upon the
first aeration of the lungs, and thus to be
associated with the changes of pressure and
flow described above. It appears that as
the blood is diverted into the lungs because
of the lowered resistance to flow in these
organs, this diversion of flow and the ac-
companying fall in blood pressure in the
ductus permit constriction and shortening
of the ductus by action of the elastic fibers
and smooth muscle cells in its wall, and
thus lead to the obliteration of its lumen.

Volume of Blood in Placenta

Dr. Fritz Fuchs, of Copenhagen, who
spent a year in the Department recently
working with Dr. Louis B. Flexner, has
reported a study of the volume of the red
blood cells in the vessels of the placenta
(guinea pig) . The estimates were made by
using red blood corpuscles labeled with

DEPARTMENT OF EMBRYOLOGY

199

a radioactive phosphorus isotope. It was
found that late in pregnancy (57 days to
term) the placentas in a given animal con-
tain about 10 per cent of the body blood
volume. The content of the fetal blood ves-
sels of the placenta is equal to the amount
of blood on the maternal side. The red cell
volume is higher on the fetal side. The
placentas contain 11 to 18 per cent of the
total fetal blood, and 55 to 79 per cent of
the fetal blood passes through the placenta
in one minute.

Oxytocic Drugs

Dr. Reynolds, with his long experience
in the physiology of the uterus as a con-
tractile organ, continues to be called upon
by clinical associates to assist in the plan-
ning and analysis of their experimental
studies. Physicians at the New York State
College of Medicine (Cumberland Hospi-
tal), Brooklyn, namely Drs. Altman, Lu-
bin, Waltman, and Tisdall, joined with
him during the past year in the publica-
tion of two brief papers on drugs used in
obstetrical practice. One of these deals
with the safe use of pituitrin in difficult
labor, by intravenous injection. The other
discusses the action on the uterus of a
newly introduced spasmolytic drug, dehy-
droergotamine-45, which is an antagonist
of adrenaline. In this latter work the toko-
dynamometer developed by Dr. Reynolds
with the assistance of Mr. Heard, men-
tioned in several previous Year Books, was
used as the chief research instrument. The
results, which are chiefly of clinical impor-
tance, need not be discussed here.

Biochemistry and Physiology of
Uterine Muscle

Results thus far achieved in the intensive
long-term investigation of the biochemis-
try and physiology of uterine muscle, led
by Dr. Csapo, were summarized in the

Ciba Foundation Lecture delivered by Dr.
Corner on behalf of Dr. Csapo and himself
at London in December 1952 and pub-
lished in March 1953 (see bibliography).
Some of the findings described in this lec-
ture, and in two papers by Csapo and Cor-
ner, have received preliminary mention in
previous Year Books, especially No. 51
(1951-1952), pages 166-167. E* r - Csapo
proved several years ago that the well
known contractile protein complex of vol-
untary muscular tissue, known as actomyo-
sin, exists also in the involuntary uterine
muscle. He has now proved that this sub-
stance is the specific contractile material
in uterine muscle. Having discovered that
the concentration of actomyosin is con-
trolled by the ovarian estrogenic hor-
mone, Dr. Csapo obviously had available a
method for altering the amount present at
a given time: that is to say, by administra-
tion of graded doses of estrogen to cas-
trated female rabbits, thus building up
the uterine tissue to any required degree.
When this was done, it was found that the
work which the muscle can do under stim-
ulation is strictly proportional to the con-
centration of actomyosin in the tissue.

Dr. Csapo, Dr. J. Gergely, and Dr. J. H.
Menkes have shown that not only the con-
tractile protein (actomyosin), but also the
energy-rich substance which provides fuel
for the contractions (adenosine triphos-
phate, "ATP") is controlled by the estro-
genic hormone. The amounts of the two
substances run parallel, probably by reason
of a quantitative molecular (stoichometric)
relation between them. The rate of forma-
tion of one of them is evidently regulated
by the other; presumably the myosin com-
ponent of actomyosin, acting as an enzyme
(ATP-ase), is involved in this relation.

The accuracy of the above-stated conclu-
sion that actomyosin is the specific contrac-
tile agent, and the quantitative value of

200 CARNEGIE INSTITUTION OF WASHINGTON

what follows in this summary, depend on behavior of uterine muscle in vitro. Pre-
the use of isometric rather than isotonic liminary experiments indicated that under
measurement of uterine muscle activity. A estrogen domination the resting uterus con-
fuller theoretical explanation of this point, tains more intracellular potassium than is
and a discussion of its implications, which necessary for the development of maxi-
are now being explored, will be found ear- mum tension, and therefore the muscle has
lier in this report, under "Research in to be stimulated frequently to make it lose
Progress." intracellular potassium and thus develop

Early in the course of preliminary ex- maximum tension. On the contrary, the
periments on the recording of isometric uterus under progesterone domination con-
muscle, Dr. Csapo and Dr. Corner discov- tains less potassium than is optimal for
ered that uterine muscle, like that of the maximum tension, and consequently the
heart, exhibits the "staircase phenome- more frequently it is stimulated, the less
non"; that is, the tension it exerts at a tension is developed.

single contraction depends on the length Dr. Csapo then measured the maximum

of time since the previous contraction. tension developed by strips of estrogen-

Within certain limits, the longer the inter- dominated uteri as the K:Na ratio in the

val, the less the tension developed. Using bath was increased in regular steps. It is

castrate rabbits injected with the two ovar- known from the work of others that such

ian hormones, it was then found that the treatment results in a moderately increased

characteristic slope of the staircase of uter- K:Na ratio inside the cells. Csapo found

ine muscle under the influence of estrogen that as the ratio is increased, the tension

is reversed under progesterone domination, gradually decreases, a result which agrees

It is evident therefore that the two hor- with the hypothesis that the estrogen-domi-

mones have opposing effects on some fac- nated muscle originally contained more

tor or factors intimately involved in the than the optimal intracellular concentra-

mechanism of contraction. tion of potassium. It was found that pro-

By a fortunate coincidence, at the very gesterone-dominated uteri, on the contrary,
time when the existence of a staircase ef- under similar treatment develop increased
feet in uterine muscle and its dependence tension, at least up to a certain point. This
on the ovarian hormone were thus discov- supports the belief that the uterus under
ered, Professor A. Szent-Gyorgyi and Dr. the action of this hormone has a subopti-
S. Hajdu were at work, at the Marine Bio- mal potassium concentration inside the
logical Laboratory, Woods Hole, on the cell. Decrease of the temperature at which
nature of the staircase phenomenon in frog the muscle is working gradually abolishes
heart muscle. They had formed a hypoth- the positive staircase and increases the slope
esis that the staircase is related to a system- of the reversed staircase, whereas increase
atic change in the concentration of potas- of temperature has the opposite effects,
sium in the muscle cells during the phase This hypothesis, which assumes that the
of resumed activity. For further test of this contractile power of uterine muscle in vitro
idea, uterine muscle is especially favorable is related to the potassium content of the
because of its slow contractions, its gen- muscle cells, obviously calls for further
eral controllability, and particularly the study in many directions. The continua-
reversibility of its staircase. Dr. Csapo tion studies now under way in the Depart-
therefore proceeded to study the effects of ment of Embryology are outlined above
varying concentrations of potassium on the under "Research in Progress."

DEPARTMENT OF EMBRYOLOGY

201

Technical Methods

Dr. Bent G. Boving has designed and
constructed an ingenious apparatus for ob-
serving the interior of the pregnant rabbit
uterus and the living embryos in the blas-
tocyst and implantation stages. This con-
sists of a transparent plastic chamber
within which a segment of the uterine
horn may be held practically immobile un-
der the microscope. Provision is made for
illumination and for experimental access
to the specimen under observation. With
the aid of this apparatus he has observed

and photographed the blastocyst stage of
embryonic development in situ and has
watched and experimented on the earlier
stages of the process of attachment of the
embryo to the uterine lining.

Mr. O. O. Heard has renewed his inter-
est in the role of surface forces in the cut-
ting of microscopic sections. By treatment
of the cutting facet of the knife with sub-
stances having a low coefficient of friction,
contamination of the facet with paraffin is
practically eliminated and distortion of the
section by compression is reduced. A re-
port has been prepared for press.

DIFFUSION AND APPLICATION OF RESEARCH RESULTS

Consultative services. Dr. S. R. M. Reyn-
olds devoted considerable time during the
year (two weeks in September and one or
two days monthly from October to March)
to participation in the work of a group as-
sembled by the Conservation Foundation
of New York City for the study of prob-
lems of overpopulation and the limitation
of reproductive rates. Dr. Corner, before
leaving for England, also took part in a
conference of this group.

The Department of Parasitology of the
Johns Hopkins University, School of Hy-
giene, is conducting an important research
program in vertebrate ecology with special
reference to the problems of rat infestation
in cities. To assist in the orientation of
those taking part in this investigation, four
members of the Department of Embry-
ology (Bishop, Boving, Burns, Reynolds)
led a series of seminars on the physiology
of reproduction, from October 1952 to May

*953-

Reviews. Members of the Department
were called upon, even more extensively
than in past years, to prepare reviews of
progress in their respective fields of re-
search. Some of these reviews were pre-
sented only in print, others were first de-

livered as lectures. The Ciba Foundation
Lecture of Dr. Corner and Dr. Csapo com-
prised not only the already cited report of
current research by the authors, but also a
historical review of research during the
past half century on the action of ovarian
hormones on uterine muscle.

Dr. R. K. Burns began the preparation
of a chapter on the role of sex-gland hor-
mones in the differentiation of the em-
bryonic reproductive system, for the new
(third) edition of the important handbook
Sex and Internal Secretions.

Dr. S. R. M. Reynolds published during
the year a volume, Physiological Bases of
Gynecology and Obstetrics, based on his
lectures at Montevideo, Uruguay, in 1950.
The book, which is intended for physi-
cians, presents an account of his varied re-
searches, much of which has been cited in
these annual reports during the past years.
The Scientific American for July 1952 con-
tained an illustrated semipopular review
by Dr. Reynolds of his recent work on the
umbilical cord.

Dr. David W. Bishop contributed a
chapter on the metabolism of the sperma-
tozoa to a published symposium on the
testis and ovary, eggs and sperm cells. He

202

CARNEGIE INSTITUTION OF WASHINGTON

has in press a chapter on the biology of
spermatozoa in the new edition of Sex and
Internal Secretions.

Dr. Louis B. Flexner, Research Associ-
ate, has also written a chapter for Sex and
Internal Secretions, on the physiology and
biochemistry of the placenta. A recent vol-
ume on the biology of mental health and
disease contains a chapter by Dr. Flexner
reviewing his studies on the physiological
development of the brain cortex. His Har-
vey Society lecture, cited in the bibliog-
raphy below, is also a review of aspects of
this subject.

Lectures. During his incumbency of the
George Eastman Visiting Professorship at
Oxford, Dr. Corner gave lecture courses,
during two terms, on embryology, the
physiology of reproduction, and the his-
tory of medicine. He delivered lectures
also at London (two), Cambridge, Edin-
burgh (two), Belfast (two), Copenhagen
(two), Lund, and Oslo.

Dr. Burns gave six lectures at the Johns
Hopkins Medical School, including two
talks for the Anatomy Department, three
in Urology, and one in the Women's
Clinic. He also spoke at the Armed Forces
Institute of Pathology, Washington, D. C.

Dr. Reynolds gave three lectures at the
Johns Hopkins Medical School, including
one in Pediatrics and two in Medicine. He
presented lectures and talks at the Cardio-
vascular Laboratory, U. S. Public Health
Service (Washington), the Obstetrical So-
ciety of Boston, the Brooklyn Gynecologi-
cal Society, the American College of Sur-
geons (in a panel discussion), the Research
Laboratories of Merck and Company, and
hospital staffs in several cities. He de-
livered formally designated lectures at
Wayne University and the University of
Michigan.

Dr. Csapo lectured once each for three
departments of the Johns Hopkins Medi-

cal School, namely, Medicine, Ophthalmol-
ogy, and the Women's Clinic. He took
part in a symposium on the biological role
of potassium, held at the University of
Minnesota in September 1952. His con-
tribution will be published.

Dr. Bishop gave six seminar talks at
Swarthmore College and also lectured at
the Johns Hopkins Medical School (Wom-
en's Clinic), the Johns Hopkins School of
Hygiene, and Goucher College.

Dr. Ramsey gave a course of lectures in
obstetrical and gynecological pathology at
George Washington University, School of
Medicine.

The Research Associates of the Depart-
ment were also in demand for lectures on
the work done at the laboratory or in asso-
ciation with it. Dr. C. H. Heuser ad-
dressed the Georgia Medical Society on
young human embryos, and the County
Medical Society of Spartansburg, South
Carolina, on placentation in the opossum.
Dr. Arthur T. Hertig spoke before the
New England Pathological Society and the
Pittsburgh Obstetrical and Gynecological
Society, on the pathology of very early hu-
man embryos. Dr. Louis B. Flexner's Har-
vey Society lecture has been cited above
under "Reviews." He was also Ferris Me-
morial Lecturer at Yale University, speak-
ing on the development of the cerebral
cortex.

Mr. Chester F. Reather, former depart-
mental photographer, is the author of
an article, "Photography in Embryology,"
published in the Eastman Kodak Com-
pany's periodical, Medical Radiography
and Photography. A beautifully printed
two-page chart of human and primate em-
bryonic development which accompanied
the article was reproduced for use on a
wall chart and has been widely distributed
to teachers of embryology by the Eastman

DEPARTMENT OF EMBRYOLOGY

203

Kodak Company. All the illustrations Grill, Mrs. Padget, and Miss McCarthy all

were made by Mr. Reather from specimens gave help in this matter.

in the Carnegie Collection. Scientific illustrations. At the Annual

Visiting groups. In October 1952, the Meeting of the Carnegie Institution of
Department of Embryology was honored Washington in December 1952, Mr. James
by a visit of the Canadian Gynecological F. Didusch, departmental artist since the
Visiting Club, at which a series of talks foundation of the laboratory, arranged an
and demonstrations was arranged by mem- exhibit of more than 150 drawings repre-
bers of the Department. In the spring of senting his life work as illustrator of em-
1953 a demonstration was arranged for the bryological, surgical, and gynecological re-
Junior Honors Seminar, Department of search. The exhibit included examples of
Zoology, Swarthmore College; and the the technical procedures of illustration at
senior biology class of the Bryn Mawr Pre- all steps from the original specimen to the
paratory School, Baltimore, was given a published plate. Mary E. McCarthy, as-
tour of the laboratory and a demonstration sistant modeler, also showed a panel of il-
of human embryology. lustrations. Mr. Didusch's pictures were

Motion picture film. Progress is being again shown in January and February 1953

made on a film illustrating the develop- at the Department of Embryology. The

ment of the human ear and auditory or- great appreciation which the exhibit re-

gans, to be produced under the auspices of ceived is a tribute to his unequaled skill as

the American Academy of Ophthalmology an anatomical artist and his constant striv-

and Otolaryngology, by the scientific mo- ing for accuracy and perfection of style,

tion picture firm of Sturgis and Grant. In Mr. R. D. Grill entered four photo-

this undertaking, which follows the highly graphs made by him in the course of the

successful film on the embryology of the Department's research program, in the

eye mentioned in previous Year Books, the 17th Rochester International Salon of Pho-

Department is again contributing techni- tography and the 10th Syracuse Salon. All

cal advice and the use of illustrative ma- four entries were hung in the two exhibi-

terial. Dr. Bartelmez, Mr. Heard, Mr. tions.

BIBLIOGRAPHY

Aberle, S. D. See Corner, G. W.

Altman, S. G., R. Waltman, S. Lubin, and
S. R. M. Reynolds. Oxytocic and toxic
actions of dihydroergotamine-45. Amer.
Jour. Obstet. and Gynecol., vol. 64, pp. 101-
109 (1952).

Ardran, G. M., G. S. Dawes, M. M. L. Prichard,
S. R. M. Reynolds, and D. G. Wyatt. The
effect of ventilation of the foetal lungs upon
the pulmonary circulation. Jour. Physiol.,
vol. 118, pp. 12-22 (1952).

See Reynolds, S. R. M.

Bartelmez, G. W. Factors in the variability of
the menstrual cycle. Anat. Rec, vol. 115,
pp. 101-120 (1953).

BooJ^ review: Purkyniana, by Jan Ev.

Purkyne. Anat. Rec, vol. 116, pp. 117-119

(i953)-
Bishop, D. W. The metabolic machinery of

sperm activity. In: Studies on testis and
ovary, eggs and sperm, ed. E. T. Engle, pp.
95-110. Springfield, 111., C. C. Thomas
(1952).

and A. Tyler. Reactions of mammalian

spermatozoa to egg substances. (Abstract)
Anat. Rec, vol. 113, p. 526 (1952).

Boving, B. G. Internal observation of rabbit
uterus. Science, vol. 116, pp. 211-214 (1952).

Boo\ review: Medicinens Historie, by

E. Gotfredsen. Bull. Hist. Med., vol. 27, pp.
79-82 (1953).

Chacko, A. W., and S. R. M. Reynolds. Em-
bryonic development in the human of the

204

CARNEGIE INSTITUTION OF WASHINGTON

sphincter of the ductus venosus. Anat. Rec,
vol. 115, pp. 151-173 (i953)-
Corner, G. W. The events of the primate
ovarian cycle. (21st Huxley Lecture, Char-
ing Cross Hospital, London, May 28, 1952.)
Brit. Med. Jour., vol. 2, pp. 403-409 (1952).

Attaining manhood: A doctor talks to

boys about sex and reproduction. English
edition, revised. London, Allen and Unwin

(i953)-

Attaining womanhood: A doctor talks

to girls about sex and reproduction. English
edition, revised. London, Allen and Unwin

(i953)-

and S. D. Aberle. Twenty-five years of

sex research: History of the National Re-
search Council Committee for Research in
Problems of Sex, 1922-1947. 248 pp. Phila-
delphia, W. B. Saunders Co. (1953).

and A. I. Csapo. Action of the ovarian

hormones on uterine muscle. (4th Ciba
Foundation Lecture, London, December 4,
1952.) Brit. Med. Jour., vol. 1, pp. 687-693

0953) •

See Csapo, A. I.

Csapo, A. I., and G. W. Corner. The antagonis-
tic effects of estrogen and progesterone on
the staircase phenomenon in uterine muscle.
Endocrinology, vol. 51, pp. 378-385 (1952).

The effect of estrogen on the

isometric tension of rabbit uterine strips.
Science, vol. 117, pp. 162-164 (1953).

See Corner, G. W.

Dawes, G. S. See Ardran, G. M.

Enders, R. E. Reproduction in the mink (Mus-
tela visoti). Proc. Amer. Philos. Soc, vol.
96, pp. 691-755 (1952).

Flexner, L. B. The development of the cerebral
cortex: A cytological, functional, and bio-
chemical approach. The Harvey Lectures,
ser. 47, pp. 156-179. New York, Academic
Press (1951-1952).

Physiologic development of the cortex of

the brain and its relationship to its morphol-
ogy, chemical constitution, and enzyme

systems. In: The biology of mental health
and disease, chap. 13. New York, Paul B.
Hoeber (1952).
Fuchs, F. The red cell volume of the maternal
and foetal vessels of the guinea pig placenta.
Acta physiol. scandinav., vol. 28, pp. 162-171

(i953).
Gilbert, P. W. The premandibular head cavities
of the opossum, Didelphis virginiana. (Ab-
stract) Anat. Rec, vol. 115, pp. 392-393

(i953)-
Light, F. W., Jr. See Reynolds, S. R. M.
Lubin, S., R. Waltman, L. Tisdall, and S. R. M.

Reynolds. An evaluation of intravenous

pituitrin. Amer. Jour. Obstet. and Gynecol.,

vol. 64, pp. 248-259 (1952).

See Altman, S. G.

Prichard, M. M. L. See Ardran, G. M.; Reyn-
olds, S. R. M.

Reather, C. F. Photography in embryology.
Med. Radiogr. and Photogr. (Eastman
Kodak Co.), vol. 28, pp. 12-19 ( T 95 2 )-

Reynolds, S. R. M. Physiological bases of gyne-
cology and obstetrics. Springfield, 111., C. C.
Thomas (1952).

The proportion of Wharton's jelly in the

umbilical cord in relation to distention of
the umbilical arteries and vein, with obser-
vations on the folds of Hoboken. Anat. Rec,
vol. 113, pp. 365-377 (i95 2 )-

The umbilical cord. Sci. American, vol.

187, pp. 70-74 (1952).

F. W. Light, Jr., G. M. Ardran, and

M. M. L. Prichard. The qualitative nature
of pulsatile flow in umbilical blood vessels,
with observations on flow in the aorta. Bull.
Johns Hopkins Hosp., vol. 91, pp. 83-104
(1952).

See Altman, S. G.; Ardran, G. M.;

Chacko, A. W.; Lubin, S.
Tisdall, L. See Lubin, S.
Tyler, A. See Bishop, D. W.
Waltman, R. See Altman, S. G.; Lubin, S.
Wyatt, D. G. See Ardran, G. M.

DEPARTMENT OF GENETICS

Cold Spring Harbor, Long Island, New Yor\
M. DEMEREC, Director

The most important event of the past
year was the completion of our construc-
tion program, which provided two new
laboratory buildings for this Department
and a lecture hall to be used jointly by
the Department and the Biological Lab-
oratory of the Long Island Biological
Association. This project, planned since
1946 and under construction since Au-
gust 1951, was finally completed in the
spring of 1953. Formal opening ceremo-
nies were held on May 29, and were at-
tended by representatives of the Trustees
and officers of the Carnegie Institution of
Washington, the Carnegie Corporation of
New York, and the Long Island Biologi-
cal Association; members of that Associa-
tion who live in the vicinity; a few of their
friends; and the staffs of the two labora-
tories. The group was welcomed by Mr.
Amyas Ames, President of the Long Island
Biological Association, who introduced M.
Demerec, to discuss briefly the history of
the Cold Spring Harbor laboratories, and
Dr. Vannevar Bush, the principal speaker
of the evening. The new laboratories were
open for inspection, and the research staffs
held an informal demonstration of selected
problems representing their current work.

A second outstanding event of the year
was the purchase of about twenty acres of
adjacent property from the estate of Mary
E. Jones, one-half by the Institution and
the other half by the Long Island Biologi-
cal Association. The property is well
adapted for experimental studies of a wide
variety of plants and animals in their nat-
ural habitats, and will be used for this
purpose by scientists working at the lab-
oratories. In addition, the acquisition of

this land ensures for the laboratories the
privacy that is essential for their efficient
operation.

Research

During the past year McClintock has
emphasized two aspects of the study of
mutation in maize: the types of modifica-
tion in genie expression that arise from
changes occurring at a single locus in a
chromosome, and the control of mutation
at specific loci by the genotypic constitu-
tion of the nucleus. The first of these in-
vestigations involved examination of ap-
proximately sixty independent alterations
occurring at a particular locus and result-
ing in modifications in expression of neigh-
boring genes. These alterations are initi-
ated by the transposable chromosomal
unit Ds when it occupies this particular
locus. It was found that Dominated modi-
fications could affect the action not only of
the known genetic factor immediately ad-
jacent to Ds, but also of genetic factors
known to be located four crossover units
to the left and at least two crossover units
to the right of it. The extent of spread
along the chromosome of genetic change
arising from Ds-'mduced modifications dif-
fered in different cases. The possible range
of this influence is difficult to determine,
because spread of the mutational effect
beyond the limits mentioned above incor-
porates genetic factors whose modified
action affects viability or functioning of
gametes, or induces a dominantly ex-
pressed inhibition of growth capacity. An
additional finding of this study was that
mutations which produce similar changes
in phenotypic expression of known genetic
factors may arise from dissimilar types of

205

206

CARNEGIE INSTITUTION OF WASHINGTON

modification. Such dissimilarities are de-
tected by differences in viability of homozy-
gote or heterozygote, in gamete function-
ing, in crossover frequencies in the vicinity
of the mutated factor, and in capacity for
subsequent reverse mutation. The evidence
suggests that no two mutation-inducing
events at a locus are exactly alike, regard-
less of similarities in mutational expres-
sion of the genetic factor concerned. Evi-
dence of genotypic control of mutational
type has been obtained with regard to two
of the mutable loci under investigation. In
both cases, one particular type of mutation
will occur in haploid nuclei that receive a
particular genetic constitution as a conse-
quence of meiotic segregations. In one of
these cases, McClintock has identified a
factor, located in a separate chromosome
of the complement, whose presence con-
trols the occurrence of mutation in those
haploid nuclei receiving it. She concludes
from this, in conjunction with evidence pre-
viously obtained, that particular changes in
the genetic constitution of nuclei, whether
arising from somatic or from meiotic seg-
regations, are responsible for controlling
the occurrence of mutation at a number of
mutable loci.

Studies carried on with the bacterium
Escherichia coli and dealing with the mech-
anisms of spontaneous and induced mu-
tability have been continued by Demerec,
Hanson, Labrum, Galinsky, Hemmerly,
and Berrie. It is reasonably certain that
each of the specific types of change ob-
served in these studies represents mutation
occurring either at one locus or at a very
few loci. The results of the past year's ex-
periments give additional support to our
working hypothesis regarding the origin
of genetic changes. This hypothesis postu-
lates that mutations are caused by some
disturbance of the physiological function-
ing of cells, which may occur spontane-
ously or be induced by the action of a mu-

tagen, and that each member of the gene
system of a cell reacts in its own specific
way to such a disturbance.

Demerec and Hanson have made a fur-
ther investigation of the phenomenon of
mutagen stability — the fact that the degree
of mutability of certain spontaneously mu-
tating characters cannot be increased by
mutagenic treatment. Tests made this year
with an additional assortment of muta-
genic chemicals failed to induce reverse
mutations in the five mutagen-stable nu-
tritional deficiencies in our collection. The
mutagens tested now include ultraviolet
radiation, X-radiation, and eleven chemi-
cals. Demerec and Hanson have also
found that bacteria grown in shaken cul-
tures are much less sensitive to the muta-
genic action of manganous chloride than
are bacteria grown in aerated cultures, a
finding which is in agreement with the
earlier conclusion that the degree of muta-
genic effectiveness of manganous chloride
treatment is determined by the physiologi-
cal condition of the treated bacterial cells.
A further analysis of delayed effect, the
phenomenon responsible for delay in the
appearance of induced mutants, has shown
that it is of widespread occurrence. Data
concerning induction of reverse mutations
in thirty-one cases of nutritional deficiency
showed that delayed effect was present in
seventeen cases. It was also found in the
four cases of mutation involving phage
resistance and the seven cases involving
streptomycin resistance studied so far.

Labrum has determined the pattern of
delayed appearance of induced mutants
resistant to bacteriophage T5 in bacterial
populations maintained at different gen-
eration times but at a constant tempera-
ture. His results show that, regardless of
the generation time, delayed appearance
of induced T5-resistant mutants is char-
acterized by an initial phase, during the
early cell divisions, when the mutants ap-

DEPARTMENT OF GENETICS

207

pear at a slow but gradually increasing
rate. This is followed by a second phase,
throughout which mutants appear at a
high linear rate that is constant per unit
time. Additional experimental evidence
suggests that the initial slow rate of ap-
pearance of induced mutants represents a
true genetic delay, but that irregularity in
the time of onset of cell division may be
involved in the second phase, during
which most of the induced mutants are
manifested.

The modifying effect of post-treatment
on the mutagenic action of ultraviolet ra-
diation has been studied by Galinsky and
Berrie. Both made comparisons of fre-
quencies of induced mutants in irradiated
cells exposed after treatment to tempera-
tures of 15° C and 37° C, Galinsky study-
ing mutations to resistance to phage T3,
and Berrie working with reversions from
three nutritional deficiencies. Both found
significantly lower frequencies of induced
mutants in the 15° series, indicating the
probability that the temperature effect,
which was first reported by Witkin (Year
Book No. 51, 1 951-1952, pp. 200-203), is a
general one. Berrie was further able to
determine that the temperature-sensitive
period corresponds approximately to the
first third of the first division cycle of
treated cells.

Hemmerly began investigations of the
genetic effects of thermal neutrons, by de-
termining the frequencies of revertants in
several nutritionally deficient strains ex-
posed for 48 or 72 hours in the nuclear
reactor at Brookhaven National Labora-
tory. The observed degree of mutagenic
effect was so low that the possibility of
gamma-ray contamination as the responsi-
ble factor could not be excluded.

Visconti and Symonds have investigated
an unstable stock of bacteriophage T2,
which forms sectored plaques. The sec-
tors contain r mutants that differ physio-

logically and genetically from those ap-
pearing in wild-type stocks. The sector-
ing is not the result of an excessive rate of
mutation, but has been explained in terms
of a prior mutation, which causes selection
in favor of the r mutants during plaque
formation.

Last year Visconti showed that a bacte-
rial cell can be infected twice with phage
T2, with an interval as long as 8 minutes
between the two infections, and still pro-
duce phage progeny derived from both
infections. Visconti and Garen have now
investigated the kinetics of intracellular
growth and recombination of phage under
these conditions of reinfection. They find
that the rates of growth and recombination
are normal, and that the genetic markers
introduced by the second infecting phage
appear in the progeny less than 4 minutes
after the second infection has occurred.
The data indicate that the second infect-
ing phage does not set up a separate intra-
cellular vegetative pool, but instead enters
immediately into the pool established by
the first phage.

The results of further attempts by Her-
shey, Hudis, and Chase to identify chem-
ical components of phage T2 that control
reproduction of the virus continue to point
to nucleic acid. Isolated virus does not con-
tain appreciable amounts of basic protein.
The membrane protein is metabolically in-
ert during reproduction. Synthesis of viral-
precursor nucleic acid occurs at times and
rates consistent with the idea that nucleic
acid contains the genetic determinants.

Nucleic acid isolated by Garen from
phage T2 has been subjected to light-scat-
tering measurements by Dr. E. Reich-
mann, of the Chemistry Department of
Harvard University. The measurements
indicate a molecular weight of about ten
million, which corresponds to ten mole-
cules of desoxyribose nucleic acid per
phage particle.

208 CARNEGIE INSTITUTION OF WASHINGTON

McDonald has continued her studies of position. A study of the desoxyribonucleo-

the properties of onion-root-tip desoxyri- proteins of sea urchin sperm, trout sperm,

bonuclease. Unlike the desoxyribonuclease and calf-thymus nuclei showed that each

previously crystallized from bovine pan- substance has unique and distinctive prop-

creas, this enzyme does not appear to de- erties, expressive of inherent genetic differ-

grade intracellular desoxyribonucleic acid ences. The information now available con-

unless the cells have first been treated with cerning the individual characteristics of

ribonuclease. Similar results are obtained these several desoxyribonucleoproteins pro-

with the desoxyribonuclease of calf spleen, vides a favorable background for analysis

A direct comparison of the modes of ac- of nuclear function and the mode of action

tion of the various desoxyribonucleases of mutagenic agents.

cannot be made, however, until they have A specific effect induced by nitrogen

been isolated and purified. Methods are mustard in gamete nuclei of the parasitic

now being devised for the isolation, puri- wasp Habrobracon has been studied by

fication, and crystallization of spleen des- von Borstel. Nuclear division is blocked

oxyribonuclease. at the sixth or seventh cleavage, and the

Kaufmann and Das have found that so- nuclei enlarge tremendously. Cytochemical
lutions of the enzyme ribonuclease, when analysis of the enlarged nuclei indicates
used in treatment of root-tip cells, may act that protein synthesis continues after cleav-
either as a narcotic, to inhibit mitosis tem- age has ceased, whereas desoxyribonucleic
porarily; as a mutagenic agent, to produce acid synthesis does not. This suggests that
polyploid and aneuploid chromosome com- nitrogen mustard may specifically interfere
plexes; or as a poison, to effect pycnosis with desoxyribonucleic acid synthesis; nu-
and death of the cell. The production of clei presumably stop dividing when a
mitotic and chromosomal aberrations is supply of desoxyribonucleic acid building-
associated with the degradation of ribose blocks stored in the cytoplasm is exhausted,
nucleic acid, and this relationship em- In connection with this study it was ob-
phasizes the significant role of ribose nu- served that chromosome synthesis and di-
cleic acid in determining chromosome vision can occur in the partially exuded
form and mitotic distribution. Since ribose contents of broken Habrobracon eggs,
nucleic acid is an extremely labile struc- Such eggs will provide a favorable system
tural component, it now seems probable for the study of nuclear division, in which
that many of the induced abnormalities mitosis is synchronous and there are no
effected by exposure of living cells to limiting membranes between the nucleus
chemical and physical agents are attributa- and the environment,
ble to alterations in their constituent ribo-

nucleoproteins, rather than to changes in dtaff

desoxyribonucleic acids or desoxyribonu- Evelyn M. Witkin was on leave of ab-

cleoproteins. sence this year, and took no active part in

Bernstein has found that the desoxyribo- the research of the Department. N. Vis-
nucleoproteins of cell nuclei can be ex- conti, who became a junior member of
tracted in distilled water. The physical the staff in October 1952, resigned at the
and chemical attributes of these nucleo- end of March 1953, to return to Italy,
proteins are unlike those of either nucleic The Department had two fellows, Rob-
acid or protein. They are characterized by ert C. von Borstel, who worked with
differences in solubility, viscosity, and com- Kaufmann as a Carnegie Institution Fel-

DEPARTMENT OF GENETICS

209

low, and Alan Garen, who worked with
Hershey under a fellowship from the Na-
tional Foundation for Infantile Paralysis.
Dr. von Borstel will leave, when his one-
year fellowship terminates at the end of
September, to take a research position with
the Biology Division of Oak Ridge Na-
tional Laboratory. Neville Symonds, in
this country from Britain under a fellow-
ship of the Rockefeller Foundation, has
spent about five months at the Department
working with Visconti and Hershey.

Co-operative Work

Continued close co-operation with the
members of the Biological Laboratory has
had a stimulating effect on research at the
Department. The five members of the
Laboratory staff took part in our weekly
staff meetings for discussion of current
research; and both they and their assistants
attended our weekly seminar lectures and
participated in the regular meetings for
review of current literature. Year-round
research at the Biological Laboratory is
focused on genetical problems related to
those studied at this Department. V. Bry-
son, W. Szybalski, and L. H. Geronimus
are working with bacteria; and B. Wallace
and J. C. King use Drosophila in their
studies.

Our staff members profited also by asso-
ciation with other research workers who
stayed at the Biological Laboratory for
various periods during the year. These
included the following scientists: William
Hayes, of the University of London; Frank
Fenner, of the Australian National Uni-
versity; W. Weidel, of the Max Planck-
Institut fiir Biologie, Tubingen, Germany;
J. D. Watson, of the Cavendish Labora-
tory, Cambridge, England; Max Delbrikk,
of the California Institute of Technology;
S. E. Luria, of the University of Illinois;
Mark Adams, of New York University

College of Medicine; E. D. DeLamater, of
the University of Pennsylvania School of
Medicine; E. Caspari, of Wesleyan Uni-
versity; I. Gersh, of the University of Chi-
cago School of Medicine; D. Shemin, of
the College of Physicians and Surgeons,
Columbia University; and J. Armstrong,
of the American Museum of Natural
History.

Members of our stafT have co-operated
in various ways with scientists at other
institutions, particularly members of the
Brookhaven National Laboratory at Up-
ton, New York. Kaufmann collaborated
with Dr. M. J. Moses, Mr. W. A. Higin-
botham, and Mr. R. Chase of that Labora-
tory in working out numerous problems
connected with an effort to adapt the RCA
television camera for use in cytological
research. During the summer of 1953,
McClintock grew her experimental corn
at Brookhaven in co-operation with Dr.
W. R. Singleton. Demerec and Hemmerly
exposed bacteria to thermal neutron radia-
tion in the nuclear reactor at Brookhaven.
Kaufmann also co-operated with Dr. T. F.
Anderson, of the Johnson Foundation of
the University of Pennsylvania, in elec-
tron microscope studies.

Meetings and Lectures

The Eighteenth Cold Spring Harbor
Symposium on Quantitative Biology, held
early in June under the auspices of the
Biological Laboratory, brought together
about 270 scientists interested in basic as-
pects of research with bacterial, plant, and
animal viruses. Thirteen of the partici-
pants were from abroad.

At various times throughout the year,
members of the staff were invited to dis-
cuss their research in seminar meetings
with other groups. Demerec spoke at a
joint seminar with the Department of
Zoology of Columbia University, held at

210

CARNEGIE INSTITUTION OF WASHINGTON

Cold Spring Harbor; Kaufmann spoke at
Smith College, Hofstra College, and
Brown University; and Hershey at the
University of Colorado Medical Center,
as well as the Cold Spring Harbor Sym-
posium.

Other Activities

For the eighteenth year Demerec com-
piled and edited Drosophila Information
Service, a mimeographed bulletin prepared
at this Department from the contributions
of Drosophila research workers in all parts
of the world. The directory section of the
November 1952 issue listed 124 laboratories

and 577 research workers co-operating in
the project. A similar circular for geneti-
cists working with microorganisms, Micro-
bial Genetics Bulletin, was compiled and
edited for the fourth year by Witkin.

The Drosophila Stock Center, under the
care of Mrs. G. C. Smith, sent out a total
of 1291 cultures to research and teaching
laboratories. Of these, 172 cultures were
shipped to Europe, Asia, Australia, and
South America.

The Department's library, with Mrs. H.
H. Wheeler in charge, acquired 193 books
during the year, and received 325 periodi-
cals and serial publications.

BACTERIAL GENETICS

M. Demerec, E. L. Labrum, I. Galinsky, }. Hemmerly, A. M. M. Berrie, J. Hanson,

I. Blomstrand, and Z. Demerec

In our studies of the mechanism of
spontaneous and induced mutability we
are working with specific types of genetic
change in strain B of Escherichia coli, each
of which, we are reasonably certain, repre-
sents mutations occurring at either one lo-
cus or not more than a few loci. Our earlier
work revealed several phenomena, some
of them still unique in the genetic litera-
ture. These phenomena are mutagen sta-
bility, delayed effect, mutagen specificity,
and modifications of mutagenic effective-
ness. During the past year we continued
with studies of the mechanisms responsible
for these aspects of mutational behavior.
We made further investigations of muta-
gen stability, testing our five mutagen-
stable types with nine chemicals that have
generally proved to be potent mutagens
for our material; but none of these was
effective in inducing reversions of the five
deficiency types. We found that the muta-
genic effect of manganous chloride is con-
siderably less in bacteria grown in shaken
cultures than in bacteria grown in aerated
cultures. Our studies of the delayed effect
showed that it is a widespread but not uni-

versal phenomenon, and also that it can be
detected and quantitatively studied with
the chemostat, using a continuously grow-
ing culture. Finally, we made an extensive
study of the effects of low-temperature
post-treatment on the mutagenic action of
ultraviolet radiation. Work was also be-
gun to determine whether or not mutation
can be induced in E. coli by treatment
with thermal neutrons.

In addition to the workers named above,
our group included Mrs. G. C. Smith,
who had charge of the Drosophila colony,
and Mrs. Katherine M. Main and Mrs.
Jean W. Mclntyre, who washed and steri-
lized the glassware used in work with
bacteria.

Our work received partial support from
a grant-in-aid from the American Cancer
Society on recommendation of the Com-
mittee on Growth of the National Re-
search Council.

Mutagen Stability

In our report two years ago (Year Book
No. 50, 1950-1951, pp. 184-185) we de-
scribed the finding that in certain sponta-

DEPARTMENT OF GENETICS

211

neously mutating types the rate of muta-
tion could not be increased by treatment
with mutagens. At that time we had two
mutagen-stable genes, which had been
tested with ultraviolet radiation, manga-
nous chloride, and (3-propiolactone. Last
year we mentioned the discovery of three
additional mutagen-stable genes, making a
total of five, among 35 nutritional deficien-
cies that we had investigated. During the
past year Demerec and Hanson, in experi-
ments with a number of additional muta-
genic agents, have found that none of them
is capable of increasing mutation rate in
these mutagen-stable types above the spon-
taneous level. So far, the following muta-
gens have been tested: ultraviolet rays,
X-rays, manganous chloride, ferrous chlo-
ride, (3-propiolactone, acriflavine, formal-
dehyde, hydrogen peroxide, neutral red
with light, and four aromatic nitrogen
mustard compounds, namely, NN-di-(2-
chloroethyl)-/7-toluidine (R44), NN-di-(2-
chloroethyl)-/7-anisidine (R45), 1 :3-dimeth-
anesulphonoxypropane (GT40), and 1:4-
dimethanesulphonoxybut-2-yne (GT58) .

It has been found that strains carrying
mutagen-stable genes are not themselves
immune to the effect of mutagens, as is
shown by the fact that a treatment which
does not affect a mutagen-stable deficiency
is effective in inducing mutations of other
characters in the same strain. For ex-
ample, in a strain (Sd-4-88) carrying both
streptomycin dependence and a mutagen-
stable histidine deficiency, the Sd-4 char-
acter reacts to treatment with mutagens in
the same way as does Sd-4 m other strains
that do not carry the mutagen-stable gene.
Similarly, tests for mutation to phage re-
sistance can be made in all our strains car-
rying mutagen-stable deficiencies; and
these tests, as far as they have been carried
out, show that the locus responsible for
resistance to Ti reacts to mutagens in the
same way in strains carrying a mutagen-

stable gene and in strains not carrying
such a character. It is therefore evident
that mutagen stability is a property of
certain genes rather than of certain strains.
Of the five mutagen-stable types we now
have, one (tryptophaneless 12-22) mutates
spontaneously with a frequency of about
6 X io -11 , which is the second lowest spon-
taneous-mutation frequency found among
our 37 nutritional deficiencies; whereas
another (prolineless 12-100) mutates spon-
taneously with a frequency of about 2.9 X
io~ 8 , the second highest spontaneous fre-
quency in this collection of strains. The
other three have intermediate spontaneous
mutability; histidineless 12-23 anQl niethio-
nineless-threonineless 12-66 mutate with a
frequency of 4 X io -10 , and histidineless
R4-88 with a frequency of 5.5 X io -9 . Thus
it is evident that mutagen stability in these
types is not related to their rates of spon-
taneous mutation.

Modification of Mutagenic Effective-
ness of Manganous Chloride

Our earlier studies had shown that the
degree of mutagenic effectiveness of man-
ganous chloride treatment is influenced by
several factors that affect the physiological
condition of the treated bacterial cells. We
had found the frequency of induced mu-
tants to be almost 10 times higher in bac-
teria grown before treatment in aerated
cultures, which usually reach a concentra-
tion of about 2Xio 9 /ml, than in bacteria
grown in nonaerated cultures, which usu-
ally reach a concentration of about 2X10 8
/ml. A difference in response had also
been observed between bacteria treated
with MnCl 2 in the resting stage and those
treated in the growing stage — growing
bacteria being less affected by the same
treatment.

When we moved into new laboratories
this year, an incubator room became avail-
able to us, and it appeared more con-

212

CARNEGIE INSTITUTION OF WASHINGTON

venient to aerate the growing bacterial cul-
tures by shaking than by bubbling, par-
ticularly since the experience of other
workers had indicated that bacteria in
shaken cultures reach an even higher con-
centration. Unexpected results were ob-
tained, however, in the experiments with
MnCl 2 being carried on by Demerec and
Hanson, and the difference was traced to
this substitution of shaking for aeration
in growing the cultures. A comparison
was then made of effect of MnCl 2 treat-
ment on bacteria previously grown in
tightly closed bottles in a shaker and bac-
teria previously grown in aerated cultures,
using strains representing eight nutritional
deficiencies and streptomycin dependence.
The number of induced mutations was 4
to 50 times higher in the aerated bacteria
than in the bacteria grown in the shaker.
Thus bacteria grown in tightly closed bot-
tles in a shaker, which reach a concentra-
tion even higher than that of bacteria
growing in aerated cultures, still show a
closer resemblance, in their reaction to
MnCl 2 treatment, to bacteria grown with-
out aeration, which do not reach so high
a concentration.

Occurrence of the Delayed Effect

The "delayed effect" was discovered
seven years ago, when it was noted that
ultraviolet- or X-ray-treated cells of E. colt
had to pass through ten to twelve divisions
before the total number of phage-resistant
mutants induced by the treatment could be
manifested. Since that time, it has been
found that this delay in appearance of in-
duced mutants is a widespread, although
not a universal, phenomenon. An analysis
of induced reversions of nutritional defi-
ciencies in our collection, made by Deme-
rec and Hanson, has demonstrated the de-
layed effect in 17 out of 31 cases tested, or
about 55 per cent. In the other instances,

all the induced mutants appeared after one
division. The delayed effect has also been
found in all the cases of mutation involv-
ing phage resistance (4) and streptomycin
resistance (7) that have been studied.

Certain of our observations — that the de-
layed effect is evident in only about 50 per
cent of nutritional mutations, that each
locus has its own pattern of behavior, and
that more than one pattern of behavior can
be analyzed in the same batch of treated
cells — eliminate the possibility that the de-
layed effect is caused by a multinucleate
condition of cells, or by diploidy, poly-
ploidy, multiple-strand chromosomes, com-
posite genes, or variability in the onset of
division of treated cells. On the basis
of what is known at present, there are
two possible mechanisms that could be
responsible for the delayed effect; and it
appears likely that both of them operate,
to various extents, in different cases. These
mechanisms are (1) induced instability of
a gene, which persists through several cell
divisions (a metastable condition), and
(2) phenotypic lag, that is, delay in ap-
pearance of the phenotype because of the
continued presence in the cell of some sub-
stance produced by the gene before it
mutated.

Generation Time and the
Delayed Effect

Previous studies by Demerec (Year
Book No. 51, 1951-1952, pp. 194-196) sug-
gested that the pattern of delayed appear-
ance of induced mutants is determined by
the particular genetic locus involved and
is unaffected by the kind of mutagenic
agent used. Witkin's investigation (Year
Book No. 51, pp. 200-203) of the effect of
temperature on rates of spontaneous and
induced mutation and on delayed appear-
ance of ultraviolet-induced Ti-resistant
mutants in strain B/r suggested an inti-

DEPARTMENT OF GENETICS

213

mate relation between cell division and
the rates of appearance of both spontane-
ous and induced mutants. The work of
Novick and Szilard with the chemostat
showed that the rate of spontaneous muta-
tion to phage T5 resistance in strain B/it/f
of E. coli is independent of the rate of
cell division. As a result of this finding
and Witkin's observations, Labrum in-
vestigated the pattern of delayed appear-
ance of mutants in bacterial populations
maintained at different generation times
but at the same temperature after treat-
ment with a mutagenic agent.

In these experiments the division rate of
cells was varied by two different methods.
The first was the use of chemostats and
the second the use of two different solid
media. The tryptophane-requiring strain
B/it/f was used throughout the investiga-
tion. The mutation system studied was
the change from sensitivity to resistance
to coliphage T5.

Experiments with the chemostat. The
chemostat designed by Novick and Szilard
at the University of Chicago is a continu-
ous-flow device that maintains a popula-
tion of bacteria, dividing at a constant re-
duced rate, for an indefinite period. When
one essential nutrient is present at a limit-
ing concentration in the medium, the gen-
eration time of bacteria dividing in the
chemostat can be controlled by the flow
rate of fresh medium from the storage
tank into the growth tube.

In these experiments a synthetic lactate
medium was used, with tryptophane as the
controlling growth factor. Cells treated
with the mutagenic chemical MnCl 2 were
inoculated into each of three chemostats
adjusted so that the generation time was
different for each dividing population. At
frequent intervals samples were taken
from each population and analyzed, by
the usual plating techniques, for number
of viable cells per milliliter and for fre-

quency of T5-resistant mutants. Three ex-
periments were performed at different
times, with essentially the same results.
No correction was necessary for spontane-
ous mutations, since they did not contrib-
ute significantly to the observed increase
in mutant frequency during the time re-
quired for these experiments.

In figure 1 the results of a typical experi-
ment are presented graphically. The fre-
quencies of MnCl 2 -induced T5-resistant
mutants in each of the three populations
are plotted arithmetically as a function of
generations. Each curve is characterized
by an initial portion representing a period
during which the induced mutants ap-
peared at a slow but gradually increasing
rate. This was followed by a period dur-
ing which the bulk of induced mutants
appeared at a linear rate until the total
yield of mutants was approached. From a
comparison of the slopes of the linear por-
tions of the three curves it is evident that
the rate of appearance of induced mutants
was most rapid in the population that had
the longest generation time (10 hours, 47
minutes), and slowest in the population
with the shortest generation time (1 hour,
50 minutes).

In figure 2 the data from the same exper-
iment are plotted as a function of time. In
this case it can be seen that the rates of
appearance of induced mutants are ap-
proximately the same throughout the lin-
ear portions of the three curves. The fact
that the three curves do not coincide is due
to differences in length of the periods of
initial slow rate of appearance.

Experiments with solid media. The
phage-spraying technique developed by
Demerec has proved to be an invaluable
method for the study of delayed appear-
ance of induced mutants in E. coli. There-
fore experiments were designed with a
view to confirming the essential findings
of the chemostat studies by use of the

214

CARNEGIE INSTITUTION OF WASHINGTON

oOl H3d SINVinW 1NV1SIS3U 91 OBOnaNI - lOUfll

u 1
1

—~ ■

1
t

O >w —

* \

***^ "»»^ ^V *

^"^ **S x

o**^ \

—

c
1

w.'

( «,

x» M

*"» Ql
X, 1

* N v 1 _

V |o

\ 1

1 l 1 K i

— N-

— &

10 5

txl

z
* o

u-
o

ro qc
LsJ
CO

c 2

o
o
o

IT)

o
o
o

01 U3d SJLNVinirt 1NV1SIS38 SI Q30naN

o
o
o

_ Z

|0U|fN

a ,

a,
a

tj ,

g m

a a

tt>

d p

-d

^fci g

</5 <yj

„ '+3 (U

rt CX, d
« O O

G n> <u
4-i <u rt

d _£ .a

22 d
'53 <-y -rt
<u o gj

r lAS ^

|| I

d rt -d

f; d .d

d bo c

Sol
-d 2

DEPARTMENT OF GENETICS

215

8 0l H3d SlNVin^ liWiSIS3U 91 QBOnQNI- "\OHH

H ^

<fr

2 ^

—

M-l

a3
<u

bJO

On <U

a, fi
rt .S

CSJ

M-l

£ del
ction

"4-1

(I)

J-l

O G

, 3

fr 1

<U «4H

M-l

r D

w 52

<lT

r-| rt

ie on
utants,

4-1

T3
V

.9 a

■4-> M

.2 §

ITl

«■

rt .S3

v-i oo
<U <U

b v

frequ
medi

1— 1

Effect of ge
-induced T5

4-J

B
rt

j— 1

4—1

"Hi

'—
c3
bJO

4— 1

<L>

4->

j— >

-a

B
>>

Fig.
f Mn

B
B

_ cvj

1 1

c
• E

V 2
\

•

m
10

•

N> ♦
♦ N

•

X ^***^n 1

\ _

•

• \

— ^ ^

x y

1 1

>
>

o .«

- CO

<s>

o
o

Q.
CM

O
O
O

O
O

O
O
O

O
O

o
o

CSJ

8 0l H3d SXNVinW 1W1SIS3M QJ. Q30nGNI - 3 I0NIAI

uj
o

u.
o

UJ
CD

B
a
u
rt
<u
a.

B
2 J2

b y
o

S-< M-l

B \

bJO

«— 1 M-l H

M-l
O

M-l

O rt

B 4-J

O w

u O

rt U

co C

rt -B

a c

rt 8

t! B

5 o

I a a

B
bJO

4^ T3
B U

52 T3

-a

-a
<u

u .a

4— 1 'rt

<L> 4-J

■-rt rt

a b

<L) rt

a. -

h -a

5 G

a o

rt ja

3 B

-B
B

a rt

B "C

C oj

B <n

O JQ

2l6

CARNEGIE INSTITUTION OF WASHINGTON

spray technique. The same B/it/f strain
was used, and T5-resistant clones were
scored.

The method consisted essentially in plat-
ing a known number of MnCl 2 -treated
cells onto a solid medium, allowing the
bacteria to pass through a certain number
of divisions, and then spraying the plates
with an aerosol of a suspension of phage
T5 to determine the number of T5-resist-
ant clones. The generation time was
varied by plating the MnCl 2 -treated cells
onto two different kinds of solid medium.
One was nutrient agar supplemented with
1 per cent dextrose; the other was the same
synthetic lactate medium used in the chem-
ostat experiments with tryptophane added
in a concentration of 20 ng/m\. The gen-
eration times of strain B/it/f on the two
media were 18 minutes and 55 minutes,
respectively. The temperature was 37° C
for all incubation periods. The rates of
spontaneous mutation on the two media
were determined by the same techniques.
All induced-mutant frequencies were then
corrected for spontaneous mutations.

In figure 3 the frequencies of T5-resist-
ant clones per io 8 plated bacteria are
plotted arithmetically as a function of gen-
erations. The resulting curves show the
same general characteristics as those repre-
senting the results of the chemostat experi-
ments (fig. 1) : an initial slow rate of ap-
pearance of induced mutants, followed by
an increased rate that appears to be linear
until it approaches total expression. The
slopes of the linear portions of the curves
are not the same, indicating that the rates
of appearance of induced mutants per cell
per generation were different on the two
different media. It should also be noted
that the total yield of induced mutants was
doubled when the bacteria were plated on
lactate medium. This effect of nutrition
on the total number of induced mutants
manifested was not investigated further.

In figure 4 the same data are plotted as
a function of absolute time. The length of
the lag phase, which was i l / 2 hours on
nutrient agar plus dextrose and 2 hours
on lactate agar, was deducted from all
the plotted times. The resulting curves
are again comparable with those obtained
in the chemostat experiments (fig. 2), in
that they show a constant rate of appear-
ance of induced mutants per unit time
during the period represented by the linear
portions of the curves.

It has been suggested that irregularity
in the time of onset of division of individ-
ual bacterial cells may play an important
role in the delayed appearance of induced
mutants. An experiment suggested by Dr.
Aaron Novick was performed to test this
point. MnCl 2 -treated B/it/f cells were
inoculated into a flask of sterile, liquid
lactate medium supplemented with tryp-
tophane (20 ug/ml), so that there was ap-
proximately 1 cell per 3 ml of medium.
One ml of the inoculated medium was dis-
pensed into each of 295 glass tubes, which
were incubated for 20 hours. After the in-
cubation period a pour plate with nutrient
agar was prepared from the contents of
each tube, and the plates were incubated
for 48 hours. At the end of that time there
were 201 plates without colonies and 94
plates with one or more colonies. Thus it
appeared probable that most of the cul-
tures showing growth had started with a
single cell. On the basis of colony counts,
the cultures were grouped according to
the approximate number of divisions ac-
complished in each (table 1).

Since the lag phase for B/it/f in liquid
lactate had previously been found to be
about 10 hours, and the generation time
1 hour, the predicted number of genera-
tions after a 20-hour incubation period was
ten. The results shown in the table indi-
cate that about 72 per cent of the cultures
had passed through eight to twelve divi-

DEPARTMENT OF GENETICS

217

sions. In other words, a majority of the
cells had begun to divide at the expected
time (plus or minus two generation times) .
Further experiments of this type, using
cultures started from single cells, are neces-
sary to define more clearly the relation be-
tween differences in time of onset of divi-
sion of individual cells and delayed appear-
ance of induced mutants.

From the results of this investigation it
was concluded that the delayed appearance

TABLE 1

Distribution of single-cell cultures accord-
ing to number of divisions accomplished

IN 20 HOURS
Generations No. cultures

1 3

2 1

3 °

4 2

5 3

6 4

7 7

8 22

9 J 5

10 18

11 10

12 3

of induced T5-resistant mutants in strain
B/it/f of E. coli is characterized by two
distinct phases. The first phase, in which
mutants appear at a slow rate, occurs dur-
ing the early cell divisions; it is followed
by a second phase, in which induced mu-
tants appear at a high linear rate that is
constant per unit time. The results of the
experiment using cultures started from
single cells suggest that the initial phase
represents a true genetic lag, which can-
not be accounted for by differences in the
time of onset of cell division, and that the
length of the lag period depends at least
partly on the generation time. The second

phase, however, during which the bulk of
induced mutants is manifested at a linear
rate, may possibly be accounted for by the
finding that there was a four-generation
range in number of divisions completed, in
a majority of the cultures started from
single cells.

Effects of Temperature on Spontaneous
and Induced Mutation

Last year (see Year Book No. 51, 1951-
1952, pp. 198-203) Witkin reported that
post-treatment with low temperature fol-
lowing ultraviolet treatment reduced the
frequency of mutants resistant to phage Ti
induced by the irradiation. The frequency
of spontaneous mutants was not affected
by the cold treatment.

Galinsky this year undertook a series
of similar experiments, testing the effects
of exposure to the two temperatures, 37° C
and 15 ° C, on both spontaneous and in-
duced mutability to resistance to phage T3.
In the induced-mutation experiments, bac-
teria were treated with either ultraviolet
light or MnCl 2 , plated, and then incubated
at either 15 ° or 37°.

The frequency of spontaneously occur-
ring T3-resistant mutants on plates ex-
posed to 15 ° was found to be about 30 per
cent of the frequency on those kept at 37°.
This differs from Witkin's results with
Ti-resistant spontaneous mutants, which
showed no variation in frequency at the
two temperatures.

The effect of temperature on ultraviolet-
induced mutation to T3 resistance is as
follows. The final yield of ultraviolet-in-
duced mutations in material incubated at
37° after irradiation is about 400,000 per
io 8 cells surviving treatment, and is com-
pletely manifested in about nine genera-
tions. The final yield in material exposed
to 15 ° is about 300,000 per io 8 , and is
reached in about eleven generations. When

2i8 CARNEGIE INSTITUTION OF WASHINGTON

the results are plotted as percentages of the act differently to the same post-treatment

final yield of induced mutations as a func- condition.

tion of post-irradiation divisions at 37° and Differential response to the mutagenic
at 15 °, the two curves run parallel for action of ultraviolet light in bacteria ex-
most of their length. The 15 ° curve, how- posed after irradiation to temperatures of
ever, has a longer toe portion than the 37° 15 ° C and 37° C has been studied further
curve. This may account for the fact that by Berrie. The mutation systems utilized
with 15 ° post-treatment it takes almost by him were reversion from amino acid
two generations longer for the total num- deficiency (leucineless 12-57, and proline-
ber of induced mutants to be manifested, less WP-14) and reversion from strepto-

Treatment with MnCl 2 and subsequent mycin dependence (Sd-4). Other defi-

incubation at 37° results in a total yield of ciency types were subjected to preliminary

about 500,000 T3~resistant mutants per io 8 analysis, and all showed indications of

survivors, reached in about nine genera- responding in the same fashion as 12-57,

tions. If the bacteria are exposed to 15 ° WP-14, and Sd-4, DUt tnev did not have

temperature after MnCl 2 treatment, then suitable properties for rigorous experi-

a final number of about 150,000 mutants mental work.

per io 8 is produced, also in about nine Once it had been established that the
generations. When the results are plotted temperature effect is of widespread occur-
on semi-logarithmic paper in terms of rence, the relation between ultraviolet dose
logarithms of induced mutants per io 8 and induced mutation, with 15 ° and 37°
viable bacteria against number of divisions, post-treatment, was determined for strains
the two curves are superimposed until 12-57, WP-14, an d Sd-4. The procedure
about the fifth generation, after which they was standard for all three strains, and con-
separate to their respective levels. The sisted in growing a culture in aerated broth
end-point number at 15° is about 30 per (with the addition of 10 \ig of strepto-
cent of the yield at 37°. mycin per milliliter in the case of Sd-4)

These experiments have shown that low- until the cells were in a resting state, wash-
temperature treatment decreases the num- ing with saline, suspending in saline, ir-
ber of mutations, both spontaneous and in- radiating with a known dose of ultra-
duced, to resistance to phage T3, and also violet at room temperature, and plating
that the temperature effect is different in onto a medium that would select the in-
ultraviolet-treated and MnCl 2 -treated bac- duced revertants. The plates were pre-
teria. In the former, 15 ° post-treatment heated and precooled. Low-temperature
brings about a comparatively small reduc- treatment consisted in keeping the plated
tion in the final number of mutants, but material at a temperature of 15 ° for 30
a much slower rate of appearance during hours before allowing it to complete
the first five divisions; whereas in MnCl 2 - growth at 37°.

treated material the cold post-treatment Table 2 shows the frequencies of mu-
effects a decided reduction in the final tants induced in the three strains by ultra-
number of mutants, although there is no violet doses in the range 100 to 700 ergs/
change in the numbers manifested per gen- mm 2 with post-treatment at the two tem-
eration during the first five divisions. This peratures. In all three strains the tempera-
seems to indicate that the mutagenic proc- ture effect is manifested over most of this
esses resulting from ultraviolet and MnCl 2 range of dosage. In strain 12-57, t ^ ie e ^ ect
treatment are not identical, since they re- ceases at about 600 ergs/mm 2 ; in the other

DEPARTMENT OF GENETICS

219

two, low-temperature treatment after ir-
radiation with the higher doses is less
effective in reducing the frequency of in-
duced mutants. With doses of more than
700 ergs/mm 2 , the high degree of killing
makes the results less significant; but we
have fairly good evidence from a number
of experiments that the temperature effect
is not present at these high doses. Witkin
was able to demonstrate an effect at 800
ergs/mm 2 , using the B/r strain of E. coli;
but this result may be explained by the
fact that her material was grown in syn-

different; for example, a dose of 100 ergs/
mm 2 produces only 40 mutants per io 8
survivors in leucineless 12-57, ^ ut 54^° P er
io 8 in prolineless WP-14. When cells are
kept at 15 ° for 30 hours after irradiation,
a smaller number of reversions is induced
for any given ultraviolet dose, except in
leucineless 12-57 at tne highest doses used.
An attempt was made to identify the
period of the growth cycle during which
temperature can affect the mutagenic ac-
tion of ultraviolet. The general procedure
was similar to that already described; irra-

TABLE 2

Frequencies of mutants per io 8 survivors induced by given doses of ultraviolet radiation
followed by exposure to temperatures of 15° c and 37° c

Dose (ergs/mm 2 )

Strain Post-treatment

100

12-57 15° C 32'

37° C 40

Sd-4 15° C 468

37° C 1,303

WP-14 15° C 779

37° C 5,480

200

300

400

500

600

700

152

528

1,584

3,590

121

269

472

786

1,342

1,922

836

1,233

1,076

3,030

5,213

6,543

,980

3,803

4,673

7,216

8,913

15,900

1,573 1,605

1,577

4,463

8,440 22,300

8,154 10,980 17,930 26,000 34,000 31,000

thetic medium before irradiation. Galin-
sky has found that the tolerance of B/r
to ultraviolet radiation is greatly increased
if the bacteria are grown in synthetic
medium instead of nutrient broth. For
the strains used here, synthetic medium is
not suitable.

Within the range of doses used in these
experiments there is a proportional increase
in frequency of induced mutants as the
ultraviolet dose increases, but the manner
of increase is different when irradiated
cells are kept at 15 ° rather than at 37°.
With 37° incubation, the frequency of in-
duced mutants increases almost linearly up
to about 600 ergs/mm 2 , and then levels off.
This linear increase is characteristic of all
the strains, but the quantitative response is

diation was at a standard dose of 400 ergs/
mm 2 , and the length of time the cells
were exposed to 15 ° before transfer to 37°
was much shorter. From these experi-
ments it was found that low temperature
can lower the frequency of induced mu-
tants if the cells are exposed to 15 ° for 10
hours. This is about one-third of the lag
phase.

The temperature-sensitive period at 37°
was determined by keeping irradiated cells
for short intervals at 37° before transfer to
15 ° for 30 hours. (These were subse-
quently incubated at 37° for four days to
complete growth.) The length of the tem-
perature-sensitive period is about i l / 2 hours,
or, again, one-third of the lag phase. Thus,
although the temperature-sensitive period

220 CARNEGIE INSTITUTION OF WASHINGTON

is of different duration at 37° and at 15°, extremely effective in inducing chromo-

it amounts to the same fraction of the somal aberrations in higher organisms,

lag phase at the different temperatures. These neutrons have very low energy and

Further experiments indicated that this velocity; and their biological activity is

temperature-sensitive period corresponds to largely due to their capture by atoms in the

the first part of the lag phase. exposed cells, and the subsequent emission

It was more convenient to carry out of gamma rays, protons, and alpha rays,
some of the experiments with liquid cul- In order to determine whether thermal
tures; and it was found that cells kept in neutrons would induce mutations in E.
synthetic medium showed a 75 per cent coli, Hemmerly studied reversions to non-
lower frequency of induced mutants than deficiency in a number of nutritionally
those kept in the same medium enriched deficient strains exposed to this radiation,
with 0.01 per cent nutrient broth. This Five of the deficiencies were mutagen
phenomenon cannot be evaluated until stable; that is, it had previously been ascer-
more critical studies have been made. tained that their rate of reversion to nutri-

The existence of the temperature effect, tional independence could not be increased

in itself, is sufficient to suggest that ultra- over the spontaneous rate by treatment

violet radiation acts indirectly to produce with mutagenic agents,

mutation. Further evidence is provided by The strains to be tested were grown in

the shape of dose-mutation curves based broth, washed, and resuspended in a syn-

on the data for 15 ° and 37° post-treatment thetic medium lacking glucose. Two-ml

shown in table 2. They lead us to conclude samples of the washed bacterial suspension

that more than one reaction is responsible were placed in cellulose nitrate tubes with

for the final mutation. The fact that the Lucite caps. They were then exposed to

37° curve is approximately linear might radiation in the thermal column of the

indicate a zero-order reaction, or a one-hit nuclear reactor at Brookhaven National

phenomenon, with only one participant in Laboratory for various periods of time,

the reaction. The shape of the 15 ° dose- Control tubes were kept under the same

mutation curve is that expected from a conditions except for the exposure to ra-

reaction of higher order, in which more diation.

than one reactant participates. The line- Extremely long periods of exposure
arity of the 37° curve probably reflects a were requ ired to obtain any mutagenic
high degree of integration among the ef j ect> After ^ hours in the thermal col-
various reactions that proceed in ultra- umn? WP-14, the most sensitive strain,
violet mutagenesis. At 15 ° this integration showed a frequency of induced mutants
is upset, and the yield of mutants depends o£ about 2000 per I0 s surv i v i ng bacteria,
on the rate of the process most affected. Th i s i s l 0W er than the frequency obtained
The studies just described have produced after a dose f only I00 er gs/mm 2 of ultra-
no evidence about the nature of these in- violet lig ht. Two strains, M-3 and Sd-4-55,
termediate reactions. were tested after 72 hours of exposure.

Only the streptomycin-dependence locus of

Studies with Thermal Neutrons strain sd _ 4 „ 55 sn0W ed an increase in in-

Relatively little has been done to study duced mutation. There was no increase in

the genetic effects of thermal neutrons, induced reversion from tryptophane defi-

although it has been shown that they are ciency in the same strain. The five muta-

DEPARTMENT OF GENETICS

221

gen-stable deficiencies showed no response
to the irradiation, regardless of the length
of time of exposure.

The long period of exposure necessary
to produce any appreciable mutagenic ef-
fect, and the low degree of this effect, sug-
gest the possibility that the thermal neu-
trons themselves do not act as a mutagenic
agent. One possible source of mutagenic
activity is gamma-ray contamination in
the thermal column. This contamination
amounts to about 50 r per hour. It is
planned to treat these strains with small
doses of X-rays in order to determine
whether the extent of induction of muta-

tions observed in the experiments can be ac-
counted for on the basis of the contaminat-
ing radiation.

At present it cannot be definitely con-
cluded that the induced mutants observed
were the result of thermal neutron irradia-
tion. Regardless of the nature of the
mutagenic agent, however, it should be
noted that each gene responded to it in a
specific manner. These results agree with
the observations of Demerec, Hanson, and
Monsees (Year Book No. 51) regarding
the specific responses of nutritionally defi-
cient strains to treatment with ultraviolet
radiation, MnCl 2 , and (3-propiolactone.

GENETIC STUDIES WITH BACTERIOPHAGE T2

N. Visconti, A. Garen, and N. Symonds

Unity of the Vegetative Pool in Phage-
Infected Bacteria

A considerable body of genetic and
chemical facts has led to the following in-
terpretation of the growth cycle of phage
T2. On infecting a bacterial cell, a phage
particle is transformed into a noninfective
intracellular form called vegetative phage.
This vegetative form can multiply and, if
a bacterium has been infected with a mix-
ture of two or more related phages that
differ in their genetic composition, can also
undergo genetic recombination. During
the first few minutes after infection, called
the eclipse period, only vegetative particles
are present within the bacterium. After-
ward, infective (or mature) particles be-
gin to form and accumulate at a constant
linear rate until the bacterium lyses, at
which time several hundred mature phage
progeny are liberated. The vegetative par-
ticles are pictured as forming an intracel-
lular pool from which units are removed
at random and transformed irreversibly
into mature phage particles, which then
remain inert with respect to the multipli-
cation and recombination processes of the

vegetative particles. Thus the three critical
processes that are assumed to be operating
during phage growth are multiplication,
recombination, and maturation of vegeta-
tive particles.

Recently Visconti found that it is pos-
sible to reinfect a cell as late as 8 minutes
after the primary infection if a very high
dose of reinfecting phage (around 1500
particles per cell) is used. Using phage
T2 labeled with different genetic markers
for the two infections, he showed that the
reinfecting phage type was represented in
the progeny, and that all the recombinant
types were also present in the proportions
to be expected in mixed infections with
unequal amounts of parental phage. This
discovery leads to an additional question
about the properties of the intracellular
pool of vegetative particles, namely,
whether or not the late-infecting phage
will enter immediately into the vegetative
pool already established by the first phage.
To answer this question, Visconti and
Garen determined the rates of formation
of mature phage T2 when a bacterium is
subjected to two successive infections, with

222

CARNEGIE INSTITUTION OF WASHINGTON

an interval of 6 or 8 minutes between in-
fections. Since phages labeled with differ-
ent genetic markers at two loci were used
(T2rk for the first infection, T2r + h + for
the second), it was possible to follow indi-
vidually the formation of phage progeny
derived from both infections. Our data
reveal the following facts:

1. Reinfection of a bacterium after 6 or
8 minutes does not interfere with the nor-
mal growth processes of the first infecting
phage.

2. The genetic markers introduced by
the reinfecting phage appear in the prog-
eny (both as parental and as complemen-
tary recombinant types) at the end of the
normal eclipse period of 12 minutes, and
thereafter accumulate at the normal rate.
This is also true when the reinfecting
phage has been inactivated by ultraviolet
light. Since reinfection can occur as late
as 8 minutes after the first infection, we
have the phenomenon of formation of new
progeny less than 4 minutes after the pa-
rental phage has infected a bacterium.

3. The complementary recombinant
types (r + h and rh + ) are produced in equal
numbers throughout the period when ma-
ture progeny appear. Also, the ratio of
recombinant to parental-type progeny dur-
ing this period is the expected one for
mixed infections in which both parents
can multiply and recombine on an equal
basis. Under the conditions described, the
second phage behaves as the minority par-
ent in most of the bacteria.

These results are readily interpreted in
terms of an intracellular pool of vegetative
phage particles. We picture the first infect-
ing phage as establishing the pool. The
reinfecting phage then finds the pool avail-
able and enters immediately. Since the
pool has probably been partly filled by the
multiplication of the first phage before re-
infection, and only a few of the reinfecting
particles are able to enter the cell, the

second phage is in the minority. But once
it has entered the pool, the second phage
can participate in the multiplication and
recombination processes of the vegetative
particles on an equal basis with the first
phage. Consequently, the events that fol-
low reinfection are the same as if the two
infections had occurred simultaneously
with unequal numbers of the two parental
types. Thus we conclude that reinfecting
phage (either viable or ultraviolet-inac-
tivated) shares a common vegetative pool
with the primary infecting phage, and that
the time constants of recombination and
maturation are characteristic of the pool,
not of individual virus particles.

A New Mutant of Bacteriophage T2

Two different classes of phage that give
rise to mixed plaques have been described
by Hershey and Chase. The first of these
contains the heterozygotes, which have
been extensively studied. A phage particle
heterozygous with respect to the r (rapidly
lysing) character, for example, yields equal
numbers of r and r + progeny, and there-
fore produces a visibly mottled plaque.
Mixed plaques from the second type of un-
stable particle are sectored rather than
mottled. They contain unstable particles
like the parent, and stable r particles, but
no r + . An example of this second type of
unstable particle has now been studied by
Visconti and Symonds. Their results can
be interpreted in terms of two genetic loci.
One locus is the site of r mutations in the
unstable particles; this locus is called R.
The other locus is responsible for the sec-
toring of the plaques, in a manner partly
elucidated below; this locus is called u.
The stock producing the sectored plaques
is called R + u.

Contrary to expectation, the rate of mu-
tation of R^u to Ru is not excessive. As
measured from the frequency of occur-

DEPARTMENT OF GENETICS

223

rence in single bursts, it is about one in 10 4
generations. Consistently with this normal
rate, Ru mutants do not accumulate un-
duly during the cultivation of R + u in
broth. The sectoring of R + u plaques in
agar is therefore due to conditions peculiar
to the agar medium on the one hand, and
to the presence in the phage of the u gene
on the other.

The R locus, that is, the R character in
Ru mutants, is closely linked to the previ-
ously known locus n, with a recombina-
tion frequency of 0.3 per cent. Since sim-
ilar mutants are not found in the wild-

type stocks with appreciable frequency (n
itself is unique), one function of the u
locus is to cause the selection of a unique
class of r mutants in agar.

The loci R and u are themselves closely
linked, with a recombination frequency of
about 2 per cent.

The sectoring of R^u plaques must be
due to selection (in agar but not in broth)
in favor of the Ru mutant. Whether this
unique behavior should be attributed to
the R or the u character has not been de-
termined, and the precise role of the u
locus remains to be investigated.

ROLE OF DESOXYRIBOSE NUCLEIC ACID IN BACTERIOPHAGE

INFECTION

A. D. Hershey, June Dixon Hudis, and Martha Chase

Our report last year summarized evi-
dence that infection of Escherichia coli by
phage T2 starts off with the injection of
the viral nucleic acid into the cell, leaving
most of the viral protein on the cell surface
in metabolically inert form. This situation
seemed to offer a unique opportunity to
learn something about the function of nu-
cleic acid in viral infection, and, more
specifically, to inquire into the chemical
basis of viral inheritance. At the outset it
was clear that this project would call for
a number of different lines of investigation
into the chemistry of viral growth. During
the current year we have made an appre-
ciable start along two of these lines.

The first question we have considered is
whether any protein is injected into the
cell along with the viral desoxyribose nu-
cleic acid (DNA). This question calls for
the use of isotopically labeled phage. We
have prepared phage uniformly labeled by
growth in a medium containing Cizf-glu-
cose of high specific activity. Our experi-
ments with this material have been explora-
tory, and only provisional conclusions can

be drawn from them. They were of three
types :

1. Ci4-labeled phage was allowed to at-
tach to sensitive bacteria, and the suspen-
sion was spun in a Waring Blendor for 2.5
minutes, and then separated into cellular
and extracellular fractions by centrifuga-
tion. The infected cells remained compe-
tent to yield phage. Both fractions were
then analyzed for labeled amino acids,
purines, and pyrimidines by paper chro-
matography and radio assay. It was found
that the cellular fraction contained 80 per
cent of each of the purines and pyrimidines
contained in the whole phage, but only 20
per cent of each of several amino acids, in-
cluding lysine and arginine.

The proper interpretation of this result,
which is in agreement with the earlier
work with phage labeled with P32 and S35,
is as follows. The phage particles attach to
the bacteria by the ends of their tails. All
or most of the nucleic acid of most of the
particles then passes into the bacterial cells,
presumably through the tails of the par-
ticles. After this happens it is possible to

224

CARNEGIE INSTITUTION OF WASHINGTON

strip the empty viral membranes from the is, the protein content of the complex is
cells, by spinning in the Waring Blendor, appreciably less than 5 per cent by weight,
without affecting the subsequent course of Additional experiments of the same type
the infection. This treatment apparently with S35-labeled phage have shown that
breaks the tails of the particles, leaving the membrane protein content of the super-
stubs attached to the cells, as judged by the natant fluid can be reduced to about 0.5 per
properties and microscopic appearance cent by repeated centrifugation at 30,000 g.
(Levinthal) of the material stripped off. This suggests that the 5 per cent of protein
The stubs of the tails presumably account analyzed above consisted of fragments of
for the 20 per cent of protein, resembling membranes, and shows that more sensi-
tive whole viral membranes in amino acid tive tests for basic protein will be feasible
composition, left on the cells after strip- if this source of contamination is reduced,
ping. It is evident, however, that this type The apparently very low basic protein
of experiment cannot detect any protein content of resting particles of T2 might be
differing in composition from the mem- interpreted as an extreme example of the
branes and comprising appreciably less rule deduced by Mirsky and his collabora-
than 20 per cent of the total viral protein, tors from analyses of chromosomes of dif-
Nor can it reveal anything about the func- ferent types of cells, namely, that the con-
tion of the protein not removed by strip- tent of basic protein is correlated with
ping. different levels of metabolic activity of the

2. In cells, and particularly in chromo- tissue of origin,

somes of cells, desoxypentose nucleic acid 3. If all or part of the 20 per cent of viral

is supposed to be associated with basic pro- protein that remains attached to infected

teins rich in arginine or lysine. If this is cells after stripping in the Waring Blendor

true of the nucleic acid of T2, it should be has any metabolic function in viral growth,

possible to detect nonmembrane protein it might be expected that some of the

after disrupting the phage particles by os- parental amino acids would be incorpo-

motic shock. rated into the viral progeny, as is true of

We subjected Ci4-labeled T2 to osmotic the constituents of the parental nucleic

shock, and digested the nucleic acid with acid. Preliminary experiments by us, by

the enzyme desoxyribonuclease. Most of Kozloff, and by French have failed to

the membrane protein was then removed, demonstrate transfer of amino acids, but

in one experiment by centrifugation at the best use has not yet been made of the

30,000 g, in another experiment by precip- available techniques.

itation with antiphage. In both experi- The analyses described above confirm,

ments the supernatant fluids contained on the whole, the conjecture that viral

about 5 per cent of the total viral protein, nucleic acid rules viral reproduction, and

and the unprecipitated fraction did not dif- hence viral inheritance. At the same time

fer in amino acid composition from the re- it is clear that static analysis will not lead

mainder. The supernatant fluids also con- to any really decisive proof of this idea,

tained nearly all the viral nucleic acid (in As a second approach, we have planned a

degraded form), and one or both should comparative study of the times and rates

have contained any protein originally asso- of synthesis of viral protein and nucleic

ciated with the nucleic acid. We conclude acid during viral growth in infected cells,

that the viral nucleic acid is not associated The analysis of protein and nucleic acid

with a characteristic protein, or that, if it synthesis represents two major projects,

DEPARTMENT OF GENETICS

225

which need to be co-ordinated not only
with each other but also with related work
already under way in other laboratories,
particularly the immunological studies of
viral growth being carried on by Luria and
his collaborators. At the present time, only
the work on nucleic acid synthesis has
progressed very far. Accordingly, we shall
summarize the results of this work without
saying much about its bearing on the larger
questions.

The questions we have asked about nu-
cleic acid synthesis are very simple. We
know from Doermann's work that infec-
tive phage particles begin to form in the
infected cell about 10 minutes after infec-
tion, and that some masked form of the
virus has already multiplied to a consider-
able extent before this time. We asked
how much nucleic acid is formed during
this period of blind multiplication, and
whether such nucleic acid is in fact the
nucleic acid of future virus particles.

One method of attack depended on the
fortunate circumstance that the nucleic
acid of T2 contains 5-hydroxymethyl cyto-
sine, whereas the desoxyribonucleic acid of
the bacterial host contains only the usual
purines and pyrimidines. Making use of
these facts, we were able to work out a
quantitative chromatographic method of
analysis for the two kinds of nucleic acid
in infected bacteria. By this method we
found that the bacterial nucleic acid dis-
appears during the first 25 minutes of
viral growth, and that viral nucleic acid
begins to accumulate promptly after in-
fection. At 10 minutes, when infective
virus is just beginning to reappear, the
characteristic viral nucleic acid already
measures about 50 units (phage-particle
equivalents) per bacterium. After this
time, the synthesis of nucleic acid keeps
pace with the formation of infective par-
ticles in such a way that the cell always

contains 50 to 100 units of surplus nucleic
acid.

To determine whether this surplus nu-
cleic acid is a precursor of the infective
particles, as it should be if the hypothetical
vegetative form of the virus contains nu-
cleic acid, we made a rather thorough
study of the kinetics of DNA synthesis and
phage maturation, using P32 as a nucleic
acid label.

The experimental principles are outlined
in figure 5, which divides the source of

Sink

Preassimilated ,
postassimilated,
and parental
phosphorus

Fig. 5. Experimental arrangements for the
analysis of precursor phosphorus. P' ', phage-
precursor P32; P, phage-precursor phosphorus
(P31+P32); V ', mature-phage P32; V, mature-
phage phosphorus. (Reprinted from Journal of
General Physiology, vol. 37, 1953, p. 2.)

phosphorus for viral growth into three
parts: phosphorus assimilated by the bac-
teria before infection, which is first in-
corporated into bacterial DNA (KozlofT) ;
phosphorus contained in the parental
phage; and phosphorus assimilated by the
bacteria after infection. Any one of these
three parts can be separately labeled with
P32, as was shown previously by Cohen,
and by Putnam and KozlofT.

We first labeled phosphorus assimilated
after infection, and particularly phosphorus
assimilated during the first 4 minutes after
infection. Atoms of this phosphorus were
found to be incorporated into nucleic acid
after spending an average of 8 or 9 min-

226

CARNEGIE INSTITUTION OF WASHINGTON

utes in the cell. The radioactive phos-
phorus in the newly synthesized nucleic
acid was later incorporated almost com-
pletely into infective phage particles. The
second step in the transport required an
additional 7 or 8 minutes. These times are
in good agreement with the results of Stent
and Maal0e, who found that the transport
from medium to phage required about 14
minutes.

These results showed that one of the pre-
cursors of the virus is nucleic acid synthe-
sized after infection, presumably the sur-
plus cytosineless nucleic acid previously
detected by chromatographic analysis.

We next inquired whether this surplus
nucleic acid is the sole or principal imme-
diate precursor of the nucleic acid of ma-
ture virus. This is equivalent to saying
that the precursor pool indicated in figure
5 is a unitary pool, and contains only cyto-
sineless nucleic acid. Moreover, if it con-
tains all the cytosineless nucleic acid, as
indicated by the completeness of the trans-
fer of early-assimilated phosphorus, the
size of the pool is known from measure-
ments already described: it contains 50 to
100 phage-particle equivalents of precursor
nucleic acid.

This prediction could be tested inde-
pendently of the assumption that the pre-
cursor is nucleic acid. We knew that the
precursor is converted into phage by an
irreversible process, because practically all
the early-assimilated P32 eventually ends
up in phage. Accordingly, we could define
the specific radioactivity of phosphorus in
the precursor pool (see fig. 5) by the rela-
tion dV'/dV=P'/P. According to this re-
lation, the specific activity of the precursor
pool (P'/P) can be measured by measur-
ing the specific activity (dV fdV) of a
sample that has just been converted into
mature phage.

Suppose, now, that a small amount of
P32 is admitted into the precursor pool at

the very start, by infecting the cells with
labeled phage. There is, to be sure, a com-
plication arising from the fact that not all
the parental phosphorus enters the pool;
about 60 per cent of it goes into a "sink" of
nonprecursor material. Taking this into
account, we shall expect that when the first
phage particles are formed they will receive
nucleic acid drawn from a pool containing
about 40 per cent of the parental radioac-
tive nucleic acid, diluted by about 50 units
per bacterium of newly synthesized, non-
radioactive nucleic acid. Thus the first
virus particle formed in a bacterium in-
fected with P32-labeled phage should con-
tain about 0.8 per cent as much P32 as was
present in the parental particles. A slight
extension of this reasoning permits a very
precise measure of the size of the precursor
pool by analysis of the kinetics of transfer
of parental phosphorus to progeny phage.

This analysis shows that the precursor
pool contains phosphorus equivalent to 54
to 94 phage particles per bacterium, in
agreement with direct measures of the
amount of nucleic acid synthesized after
infection. We conclude, therefore, that the
precursor pool contains all its phosphorus
in the form of nucleic acid, and contains
all the nucleic acid synthesized after infec-
tion and not yet incorporated into infective
particles.

A similar analysis of the preassimilated
bacterial phosphorus shows that it is not
present in the pool of immediate viral-
precursor nucleic acid, but enters this pool
gradually during viral growth. This result
agrees with the fact, since established by
Kozloff, that the precursor material is bac-
terial nucleic acid, and has to be converted
into cytosineless nucleic acid before it is
directly available for viral growth. We
find that this conversion is completed in
about 25 minutes.

As a whole, the tracer experiments show
that viral nucleic acid is synthesized at the

DEPARTMENT OF GENETICS

227

rate to be expected if the precursor nucleic
acid is contained in the vegetative phage
particles identified by genetic methods. It
remains to be seen whether the synthesis
of viral-precursor protein obeys the same
or different rules.

Molecular Characteristics of the DNA
Component of Phage T2

In view of the primary role played by
the viral desoxyribose nucleic acid (DNA)
in the multiplication of phage T2, the mo-
lecular characteristics of this material have
been investigated by Garen in collabora-
tion with Dr. E. Reichmann, of the Chem-
istry Department of Harvard University.
The DNA of T2 was separated from its
protein membrane by several shakings of

an aqueous suspension of the phage in the
presence of chloroform. This procedure
removes most of the sulfur-containing pro-
tein (as indicated by the residual radioac-
tivity when S35-labeled phage is used as
a tracer) and recovers more than 50 per
cent of the DNA. Preliminary measure-
ments of the isolated DNA by light-scat-
tering techniques indicate a molecular
weight of ten million. From this value we
can estimate that a mature particle of
phage T2 contains around 10 molecules of
DNA of this molecular weight. Further
experiments are now in progress, using
viscosity and rotary-diffusion techniques
in addition to light scattering, to gain ad-
ditional information about the size and
shape of the molecules.

MUTATION IN MAIZE

Barbara McClintock

The origin of instability at a number of
loci of known genie action in maize chro-
mosomes has been described in previous
Year Books, and the nature of the unstable
expression discussed. In some of these cases,
instability appeared when Ds, a trans-
posable chromosomal unit, was inserted
next to the locus. The original and the
subsequent mutations were shown to be
expressions of changes induced at the locus
by Ds. Ds is known to produce chromatin
alterations. It was suspected, therefore,
that the mutations it induces may be ac-
companied by some form of chromatin al-
teration or reorganization. If so, no two
mutation-inducing events need produce ex-
actly the same type of change at the locus,
even though the resulting modification
effects the same change in expression of
the known genetic factor. It was decided,
therefore, to examine a number of Ds-
induced mutations in order to determine
whether or not differences could be de-

tected among them. For this purpose, a
number of independently occurring muta-
tions at a selected locus were required.
The two cases, described in Year Book
No. 51 (1951-1952), in which Ds was in-
serted into the short arm of chromosome
9 just to the left of Shi were selected
because each has provided a large num-
ber of newly induced mutations to shi
(shrunken endosperm).

Origin of the Mutants

All the mutants used in this study ap-
peared in the progeny derived from crosses
of plants homozygous for C, ds, shi, and
bz, and carrying no Ac factor, to plants
having Ac and one of the following con-
stitutions: (1) I Ds Shi Bz/I Ds Shi Bz,
(2) IDs Shi Bz/I Ds Shi bz, or (3) / Ds
Shi Bz/C ds shi bz. I (inhibitor of aleu-
rone color, dominant in action to C, which
produces aleurone color) is located approx-
imately four crossover units to the left of

228

CARNEGIE INSTITUTION OF WASHINGTON

Shi. Bz (allele of bz, recessive mutant
which alters the color of the aleurone layer
and the plant tissues to a bronze shade) is
located approximately two crossover units
to the right of Shi.

The frequency of occurrence of germinal
mutations to shi among the kernels on the
ears resulting from the above-described
crosses was recorded in Year Book No. 51.
On the ears produced by the first two of
these crosses, a few completely colored, Shi
kernels were present. Such kernels were
expected for the following reasons. In the
two cases under consideration, most of the
alterations produced by Ds result in a di-
centric chromatid and a reciprocal acen-
tric fragment. The position of break in the
short arm of chromosome 9 is at the locus
of Ds, just to the left of Sh±. The acentric
fragment, carrying /, is subsequently lost
during a mitotic cycle. If such an event
occurs in a premeiotic, meiotic, or gameto-
phytic nucleus, gametes may be formed
that are deficient for a segment of chromo-
some 9 extending from the locus of Ds to
the end of the arm. Previous studies had
shown that female gametophytes having
such deficiencies are viable and functional,
and that kernels in which / is absent can
be produced from them. In the crosses de-
scribed here, such kernels should be com-
pletely colored. Plants derived from them
should have a long terminal deficiency of
the short arm of one of their chromosomes
9. As will be shown later, this proved true
of only some of these plants. Morphologi-
cally normal chromosomes 9 were present
in the others, but the action of / carried by
the chromosome received from the female
parent was inhibited.

Change in Action of Genes Located
to the Right of Ds

An analysis of plants derived from some
of the kernels carrying newly induced shi

mutants that appeared on the ears from
cross 3 listed above was given in Year
Book No. 51. The mutants appearing on
the ears from crosses 1 and 2, however, are
of particular importance, because neither
crossing over in a heterozygote nor con-
tamination could account for the mutant
phenotype. Forty-nine plants carrying
newly induced alterations at the locus of
Shi have been examined. All were derived
from kernels that were I shi in phenotype.
Sixteen were selected from the ears pro-
duced by cross 1. Ten of these kernels
were completely colorless, indicating the
absence of either Ds or Ac. In four others,
areas exhibiting either the C Bz or the
C bz phenotype were present, indicating
the presence of both Ds and Ac. In the
remaining two kernels, areas showing only
the C bz phenotype appeared, indicating
the absence or inhibition of Bz. Nine I shi
kernels were selected from the ears pro-
duced by cross 2. Five of them were com-
pletely colorless, and four had areas of
both the C Bz and the C bz phenotypes.
From the ears produced by cross 3, nine-
teen / shi kernels were selected that
showed areas of the C Bz and the C bz
phenotypes, and five additional kernels
were selected that showed areas only of
the C bz phenotype.

Subsequent study showed that in thirty-
seven of the forty -nine plants derived from
the selected I shi kernels, Bz was present
in the / chromosome and unmodified in
action. The remaining twelve plants were
all bronze {bz) in phenotype. As will be
shown later, this phenotype appeared in
some of these plants because the locus of
Bz had been altered by the same event that
had altered the Shi locus.

Comparisons between shi Mutants

Modification affecting only Shi. Simi-
larities and differences among the thirty-

DEPARTMENT OF GENETICS

229

seven cases of mutation to shi in which
the action of Bz was unaffected by the mu-
tation-inducing event will be considered
first. In all cases, the chromosome carrying
the new shi mutation was transmitted nor-
mally through the male and female game-
tophytes, and the homozygotes were viable.
Ds was present in each, and its position
was not detectably altered by the event
that produced the mutation to shi. Rever-
sions to Shx occurred in nine of the thirty-
seven cases. In two of them the frequency
of reversion was high, and in both cases it
was ^-controlled. The reversions to Shi
were not associated with loss or change in
location of Ds. The behavior of the shi
mutants differed not only with regard to
the occurrence of reversions, and to fre-
quencies of reversions when they did occur,
but also with regard to frequency of cross-
ing over between the locus of shi and the
loci of its nearest known neighbors, / and
Bz. In one case consistent results have
been obtained in tests extending over three
successive generations. When this shi is
present, crossing over between / and shi is
increased about 50 per cent, in comparison
with standard frequencies, and crossing
over between shi and Bz is increased ap-
proximately 300 per cent. In several other
cases, the mutation to shi resulted in
marked reductions in crossing over. In a
backcross test involving one such case, no
crossing over between shi and Bz was ob-
served in a total of 3156 tested gametes;
crossing over between / and shi was
slightly reduced, but that between Bz and
a marker, Wx, located to the right of Bz,
was unaltered. In all cases where a modi-
fication in crossing over was expressed, it
was confined to the vicinity of the shi
locus.

It may be concluded that the described
dissimilarities among these shi mutants
reflected dissimilarities in the primary mu-

tation-inducing events, even though all
were initiated by Ds.

Simultaneous modification of the action
of Shi and its neighboring gene, Bz. As
was stated above, twelve of the forty-nine
plants derived from / shi kernels produced
from crosses 1 to 3 exhibited the reces-
sive bronze phenotype. Four of the ker-
nels from which these plants arose had
appeared on ears produced by cross 1, three
came from ears produced by cross 2, and
five from ears produced by cross 3. (These
last five were selected because of the ap-
pearance of C bz areas in the aleurone
layer and the complete absence of any de-
tectable Bz action, and also because in four
of them the endosperm was noticeably de-
fective.) Only in the four cases derived
from cross 1 was it certain that the Shi
and Bz loci had been altered by the same
event. The female parent in this cross was
homozygous for both Shi and Bz; whereas
one of the chromosomes 9 in the female
parents used in the other two crosses car-
ried the known recessive, bz. In the re-
maining eight of the twelve plants, the
presence of this recessive in the / .^-car-
rying chromosome could account for the
bronze phenotypic expression, the mu-
tation-inducing event having altered only
the Shi locus. Evidence obtained from
subsequent study of one of the three cases
derived from cross 2 and four of the five
cases derived from cross 3 suggests, how-
ever, that a longer segment within the
chromosome had been altered, and that it
extended from the locus of Ds to or be-
yond that of Bz or its allele bz. This evi-
dence will appear in the following dis-
cussion.

One of the four bronze plants obtained
from the / shi kernels of cross 1 was com-
pletely male- and female-sterile, and there-
fore the modification responsible for the
bronze phenotype could not be examined
further. Cytological examination of the

230

CARNEGIE INSTITUTION OF WASHINGTON

other three plants revealed no structural
alteration in their chromosomes 9. Marked
differences were noted, however, in genetic
behavior. In one of them, the /-carrying
chromosome behaved as though the known
recessives, shi and bz, were present, in that
it was normally transmitted through the
male and female gametophytes, and the
kernels homozygous for it were normal in
appearance. Also, no reversions to Shi or
Bz were observed. The behavior of the
/-carrying chromosome in the other two
plants was decidedly aberrant. It suggested
that a segment of the chromosome, includ-
ing the Shi and Bz loci, was either absent
or greatly modified. In one of these cases,
transmission of the chromosome was nearly
normal through the female gametophyte
but markedly reduced through the pollen
grain. Kernels homozygous for the modi-
fied segment were not produced, and most
of the kernels heterozygous for it were ab-
normal in appearance — smaller than nor-
mal and showing few to many wrinkled
regions. In the second of these two cases,
the chromosome was transmitted at a re-
duced rate through the female gameto-
phyte and was not transmitted through the
pollen. Most of the kernels heterozygous
for this segment were defective, and in
many of them the embryo appeared to be
dead. In both cases, Ds was present in the
chromosome having the modified segment.
Although its exact position has not yet
been determined, in both cases it was
either adjacent to or a component of the
genetically modified segment. No rever-
sions to Shi or to Bz were found in either

case.

The genetic behavior of the /-carrying
chromosome in two of the three bronze
plants grown from the / shi kernels from
cross 2 was normal. It resembled that of
a chromosome carrying the known reces-
sives shi and bz. Ds was present in the
chromosome in each of these plants, and

not noticeably altered in location. The
behavior of the / chromosome in the third
plant was similar, except that somatic re-
versions to Bz occurred and they were
^-controlled. The frequency of reversion
was low, and no germinal mutations ap-
peared in tests of several thousand gametes.
No reversions to Shi were observed. In
the event that they did occur, with as low
a frequency as the Bz reversions, detection
would be difficult. Only germinal rever-
sions to Shi, or those occurring in very
early development of the kernel, can be
recognized with certainty. Ds was present
in this chromosome, and again its location
had not been altered noticeably by the
event that produced the changed pheno-
typic expressions of Shi and Bz. It was
close to shi and to the left of the mutable
bz locus.

In one of the five bronze plants derived
from I shi kernels produced by cross 3, no
aberrant behavior of the /-carrying chro-
mosome was observed. Normal transmis-
sion occurred through the male and female
gametophytes, and the kernels heterozy-
gous and homozygous for the chromosome
were normal in appearance. Ds was pres-
ent in this chromosome. No reversions to
either Shi or Bz were seen. In the four
remaining plants, the / chromosome be-
haved as if it had a modified segment, the
modification being similar to that in the
two cases described above. Here, again,
there was a reduction of gametophytic
transmission of the chromosome carrying
the modified segment. In one plant, trans-
mission through the female gametophyte
was normal, and that through the pollen
grain only slightly reduced. The hetero-
zygous kernels were normal in appear-
ance, but no homozygotes were obtained.
Another plant showed nearly normal
transmission through the female gameto-
phyte, but markedly reduced transmission
through the pollen grain. No homozy-

DEPARTMENT OF GENETICS

23 1

gous kernels were obtained, but the hetero-
zygotes were normal in appearance. The
behavior in the remaining two plants was
similar to that in the plant just described,
except that most of the kernels heterozy-
gous for the modified segment were de-
fective — small kernels, with few to many
wrinkled areas. Cytological examination
did not reveal a detectable structural modi-
fication in the short arm of chromosome 9
in any of the plants that were heterozygous
for these modified segments. Again, Ds
was present in all the chromosomes with
modified segments, and it appeared to be
located adjacent to or to be a component
of the modified segment. No somatic re-
version to either Shi or Bz was observed.
Although it is clear from the descriptions
above that alterations affecting both the
locus of Shi and that of Bz had occurred
in at least nine of these twelve cases, and
that homozygotes exhibiting the recessive
expression could be obtained from two of
them, it is not yet possible to state whether
or not deficiency was responsible for the
recessive expression, except in the one case
where reversion to Bz occurred. In that
case, deficiency of Bz is excluded. Marked
deleterious effects are produced by some of
these alterations, even in the heterozygote;
and this situation has not previously been
encountered in studies of heterozygous de-
ficiencies in maize. Although deficiency
cannot be excluded as the cause of the re-
cessive phenotypic expression in every case,
there is at present no direct evidence of it
in any one case.

Change in Action of Genes Located to
the Left of Ds

The previous sections have considered
changes in expression of known genes lo-
cated to the right of Ds, namely, Shi and
Bz. Changes in action of /, located four
crossover units to the left of Ds, may also

occur. Evidence of this has been obtained
from plants derived from the colored ker-
nels appearing on the ears of crosses 1 and
2. As was explained earlier, the appear-
ance of such colored kernels on these ears
was anticipated; and plants derived from
them were expected to have a chromosome
9 with a long terminal deficiency of the
short arm, extending from the locus of Ds
to the end of the arm. It was considered
necessary to test this expectation. Conse-
quently, some of the colored kernels were
selected and plants were grown from
them. Eighteen plants were obtained from
the colored kernels of cross 1 and two from
those of cross 2. Owing to the pressure of
other work, time was not found to ex-
amine the chromosomes of these plants.
In order to obtain seed for later examina-
tion, each plant was self-pollinated and
also crossed by plants homozygous for C,
shi, and bz. The types of kernels appear-
ing on the ears of three of the plants de-
rived from colored kernels produced by
cross 1 and one of the two plants derived
from such kernels from cross 2 were quite
unexpected. They indicated that the ap-
pearance of color in the original kernel
from which each plant arose was not due
to a terminal deficiency, but rather to some
more localized modification within chro-
mosome 9 that had affected the action of /.
Since the kernel types and the frequencies
of types were the same on the ears of all
four plants, they can be considered collec-
tively. On the self-pollinated ears, a few
completely colorless, Shi kernels appeared.
There were 87 of them among a total of
1224 kernels, and the percentages on the
four ears were 4.6, 7.7, 7.7, and 9.0. The
remaining kernels were all colored: 604
were Shi, and nearly all of these were very
lightly colored; 533 were shi, and these had
the depth of color that is produced by three
doses of C. On the ears resulting from
the backcross, no colorless kernels were

232

CARNEGIE INSTITUTION OF WASHINGTON

present. The Shi and shi classes of kernels
appeared in equal frequencies. Nearly all
the Shi kernels were lightly colored, and
nearly all the shi